*Article* **The Role of Social Support in Participation Perspectives of Caregivers of Children with Intellectual Disabilities in India and South Africa**

#### **Shakila Dada 1,\* , Kirsty Bastable <sup>1</sup> and Santoshi Halder <sup>2</sup>**


Received: 15 August 2020; Accepted: 3 September 2020; Published: 11 September 2020

**Abstract:** Caregivers are an intrinsic component of the environment of children with intellectual disabilities. However, caregivers' capacity to support children's participation may be linked to the social support that they, as caregivers, receive. Social support may increase participation, educational, psychological, medical and financial opportunities. However, there is a lack of information on social support in middle-income countries. The current study described and compared the social support of caregivers of children with intellectual disabilities by using the Family Support Survey (FSS) in India and South Africa. The different types of social support were subsequently considered in relation to participation, using the Children's Assessment of Participation and Enjoyment (CAPE). One hundred caregiver–child dyads from India and 123 from South Africa participated in this study. The data were analysed using non-parametric measures. Indian caregivers reported greater availability of more helpful support than did the South African caregivers. Social support was associated with children's participation diversity (India) and intensity (South Africa). The child-/caregiver-reported participation data showed different associations with participation. Results from this study suggest that perceived social support of caregivers differs between countries and is associated with their child's participation. These factors need to be considered when generalising results from different countries.

**Keywords:** social support; family support survey; participation; intellectual disabilities; low- and middle-income country

#### **1. Introduction**

The introduction of the International Classification of Functioning, Disability and Health (ICF) [1] and the Child and Youth Version (ICF-CY) [2] highlighted participation as a critical health outcome [3]. Furthermore, participation has been highlighted as a human right for persons with disabilities at the United Nations Convention on the Rights of Persons with Disabilities [4]. Participation is described as an important means for achieving physical, social and academic development, cultural understanding, and community inclusion. It is argued that through participation, developmental skills are practiced until an outcome of learned skills is produced [5,6].

As the field of participation is growing, however, gaps in research have emerged. In spite of the fact that participation is reported to be influenced equally by personal and environmental factors [7–9], the bulk of research on participation has focused on personal factors. The current research has provided evidence of decreased participation for children with disabilities [9–14] and specific patterns of participation associated with the type and severity of a disability [15–19]. Studies considering the impact of environmental factors are more limited in number, but as highlighted by Anaby et al. in

a scoping review on the effect of the environment on participation, family support and geographic location are facilitators of participation, while attitudes, the physical environment, policies and a lack of support serve as barriers to participation [20].

A paucity of research is also evident in relation to the effect of the income level of the country or culture on participation. Most children with disabilities in the world live in low- and middle-income countries [21], and environmental factors have been identified as an important participation-related concept. Hence, it would have been expected that research on participation in low- and middle-income countries would be common. However, the review by Anaby et al. [20], which identified 28 studies (and three reviews), all were conducted in high-income countries. A more recent scoping review by Schlebusch et al. [22] identified 78 studies on participation from low- and middle-income countries (55% conducted after the Anaby et al. [20] review). However, only 4% (*n* = 6) of these studies were from low-income countries, with 68% (*n* = 53) conducted in upper-middle-income countries. Furthermore, again only 4% (*n* = 6) of the studies in this review considered the effect of the environment on children's participation, while the remaining studies investigated participation as a process (*n* = 7), participation as an outcome (*n* = 42), child-related outcomes (*n* = 14), and the measurement of participation or related constructs (*n* = 11) [22]. All in all, there remains a lack of research on participation of children with disabilities from low- and middle-income countries, particularly in relation to environmental factors [22,23].

The importance of research on participation from low- and middle-income countries relates specifically to differences in the environment that may affect children with disabilities' participation. As indicated by Anaby et al. [20], environmental factors may function as facilitators of or barriers to the participation. Compared to their peers in high-income countries, children in low- and middle-income countries have been identified as being at greater risk from environmental influences such as poverty, reduced educational opportunities, violence and difficulty accessing healthcare [24–26]. In addition, the studies in the Anaby et al. review [20] were mostly from English-speaking countries embracing Eurocentric/western philosophies such as the U.S., the U.K., Canada, Australia and Europe [20], which see the individual as being independent from their community. This is in contrast to Afro-/Asia-centric philosophies that are founded on collectivism or see the self as inseparable from the community [27–29]. Differences in life philosophies may affect perceptions of self and disability, perceptions or availability of support, communication, and hence participation in these settings [27–29].

Within different cultural philosophies, the role of caregivers and the impact of factors such as caregiver support may affect participation. Unfortunately, limited research has been conducted in this area. Caregivers play a much greater role in finding [30,31] and facilitating [32] opportunities for participation for children with disabilities [20,33] than for children with typical development. In fact, the responsibility of ensuring that the rights of a child with intellectual disabilities are met, is reported to fall most often on caregivers [34–36]. Such responsibilities can create additional stress for caregivers and may limit their adaptability. Different forms of social support have been described as buffers for caregivers of children with disabilities to decrease stress and increase positive parenting [37,38]. When considering participation specifically, as expressed in the ICF-CY [2], "the role of the family environment and others in the immediate environment is integral to understanding participation, ... ." [2], (p. xvi). Yet, little is known about the support experienced by caregivers of children [39], particularly in Afro-/Asia-centric countries [40]. Social support specifically is a process that "arises from formal support (medical or professional) and informal sources (extended family, friends, and neighbours) around the caregiver and family" [40], (p. 160). Social support is said to be a reciprocal interaction in which caregivers feel cared for, esteemed and valued, and in which they are engaged in a system of communication and mutual responsibility [41]. Social support enables caregivers of children with disabilities to mediate the stress that they face [38,42,43] by developing resilience [44] and increasing their situational appraisal [45] and coping strategies [46]. While reductions in stress are reported to increase well-being [47], the presence of social support for caregivers and the use of

positive caregiving styles are reported to increase the quality of caregiving [48]. Nurturing the child's self-esteem can also result in better developmental outcomes for the child [40,48].

One distinct difference between Eurocentric and Afro-/Asia-centric households is the proportion of multi-generational households (both India and South Africa) [40,49]. Multigenerational households have been highlighted as able to provide resilience and growth where this might otherwise not have been possible [50–53]. The presence of older generations in the household can, however, also add to a caregiver's responsibilities. In Eurocentric cultures, help for an older generation is provided primarily when specific needs arise (for example injury or illness), and therefore multigenerational households are less common. In Afro-/Asia-centric cultures, simply "being old" is sufficient for the provision of additional support [51,52], and the provision of this support is culturally obligatory [52,53].

The influence of caregivers of children with disabilities on their child's participation is represented in the context- and environment-related constructs of the family of Participation-Related Constructs (fPRC) model [6]. In this model, caregivers constitute a key component of their child's context (part of the environment). They provide opportunities for participation, regulate the environment, and respond to their child [54]. In spite of this key role played by caregivers of children who have intellectual disabilities, only four studies [32] have made use of tools in which the caregivers reported on participation, and none of these considered factors specific to the caregiver which may affect participation [6]. From a systems perspective, the impact of caregiver factors on the participation of a child with intellectual disabilities can also be appreciated, as the influence of each level of the system on the other levels is highlighted. This perspective is supported by studies that identify the caregiver education level, income and social support structures [33,55] as factors that may have an effect on participation [33,56,57].

The final gap that has been noticed in the literature is the notion of diagnosis. The most commonly reported disability in terms of participation is cerebral palsy [20,58–63], and research on other conditions such as intellectual disabilities is sparse. Intellectual disability is a pervasive and lifelong condition in which children present significant limitations with regard to intellectual functioning and adaptive behaviour, prior to the age of 18 years [64]. In addition to the individual challenges experienced by children who have intellectual disabilities, environmental barriers may impede the achievement of human rights [34–36]. These may include a lack of opportunities for participation in education, recreation, leisure, sporting and community activities [10,65]. For children with intellectual disabilities, the combination of individual challenges and environmental barriers can result in decreased cognitive and linguistic skills, poor motor and social skills [66], social isolation and chronic health problems [32]. A systematic review of the participation of children with intellectual disabilities identified four studies that reported that children with intellectual disabilities participated to a similar extent in leisure activities, but less in social activities within the community, recreational activities, family enrichment activities and formal activities, than did their typically developing peers [32]. Other studies not included in the review indicated decreased participation in active-physical and skills-based activities [19,66,67] and a higher proportion of participation in social and recreational activities [19]. In addition, children with intellectual disabilities were noted to participate in a significantly greater number of activities at home [68], by themselves [19] or with adults, rather than with peers [10], in comparison to children with typical development. In addition, challenges in the participation of a child with an intellectual disability were found to affect not only the child, but also to place high levels of stress on the parents and family [30].

In conclusion, there is a need to describe the influence that the environmental component of caregiver support has on the participation of children with intellectual disabilities from low- and middle-income countries [33]. The current study aimed to measure, describe and compare the social support of caregivers of children with intellectual disabilities from India and South Africa, and to determine if there is an association between the social support reported by caregivers and the participation of their children as reported by caregivers and their children. India and South Africa were selected since both countries have been identified as having cultures in which households are more

commonly multigenerational. However, India is a lower-middle-income country and has a very high reported prevalence of intellectual disability (≈6%), while South Africa is an upper-middle-income country and has a lower reported prevalence (≈2.25%) of intellectual disability [69,70].

#### **2. Aims**

This study had three key aims: firstly, to describe and compare the social support reported by caregivers of children with intellectual disabilities in India and South Africa; secondly, to determine if there was any association between the demographic factors and the social support reported by caregivers; thirdly, to determine if there was any association between the social support reported by caregivers and the participation of their children with intellectual disabilities.

The first hypothesis formulated for this study was that the social support available to caregivers in India and South Africa would be different. The second hypothesis suggested that demographic factors in India and South Africa would affect social support, and the third hypothesis stated that increased perceived social support would be associated with increased participation of the children with intellectual disabilities.

#### **3. Materials and Methods**

#### *3.1. Study Design, Sampling and Participant Selection*

A comparative group design was used for this study. Purposive sampling was used in schools for children with intellectual disabilities to identify participants. In both countries, education for children with disabilities is provided in special schools which can be either government funded or private. The support provided by these schools is dependent on a range of factors including context (rural/urban), funding and fees paid by parents. Both urban and rural schools were included in this study.

Inclusion criteria for caregiver–child dyads required children to be between the ages of 6 and 21, and to have a primary diagnosis of mild to moderate intellectual disability. Caregivers were required to be literate in Bengali, English, Afrikaans, isiZulu or isiXhosa, and children were required to speak Bengali, English, Afrikaans, isiZulu or isiXhosa. If a child's home language was not the same as the language in which the Children's Assessment of Participation and Enjoyment (CAPE) [71] was to be administered at their school, then the child needed to have been schooled in the language of the CAPE [71] for at least 1<sup>1</sup> <sup>2</sup> years in order to be included in the study.

#### *3.2. Ethics*

Ethics approval for the study was obtained from the relevant ethics committees of the institutions of higher education in both countries. Permission was obtained from the appropriate departments and heads of schools or centres. In India, participants were identified in twelve schools and centres for children with disabilities. In South Africa, permission was obtained from the Department of Education in six provinces. Permission was also obtained from the principals and governing bodies of the schools identified. Eleven government schools and four private schools gave permission for their children to participate in the study.

#### *3.3. Participants*

A total of 223 caregiver–child dyads participated in the study, with 100 dyads from India and 123 from South Africa. The children had a mean age of 12:4 (years:months), and the sex composition of the sample was 61.3% male and 38.7% female. Although the reporting caregiver was primarily the child's mother (73.6%), fathers (15.6%) and other caregivers (10.8%) also reported on their children. More than half of the caregivers had at most a Grade 12 education (India 68%; South Africa 64%), and 64% of caregivers reported a household income of less than ZAR 30,000.00 (approximately EUR 1500.00) per month. Indian caregivers reported between one and six children residing in the household (M = 2), while the South African caregivers reported between one and 13 children (M = 3) in the household. Indian caregivers reported having grandparents living in the house in 67.4% of families, and other family members in 47.7% of families. South African caregivers reported having grandparents living in the house in 32.6% of families, and other family members in 52.3% of families. Statistically significant differences were evident in the demographic data of caregivers from India and South Africa.

The summarised demographic data of the participants are presented in Table 1.



Note \* the p value is significant. <sup>1</sup> Percentages may not add up to 100 due to rounding. <sup>2</sup> Pearson's chi-square *p* < 0.05. <sup>3</sup> Fisher's exact test–one-sided.

#### *3.4. Materials*

The availability of support to caregivers of children with intellectual disabilities in this study was determined using the Family Support Scale (FSS) [72]. The FSS [72] is a 19-component scale that asks caregivers to rate the helpfulness of support from various sources, for example, spouse, parents, friends, and parent groups. For each support source, the caregiver indicated on a Likert scale whether the support was not available (0), not at all helpful (1), sometimes helpful (2), generally helpful (3), very helpful (4) or extremely helpful (5). In the scoring of the FSS [72], the 19 sources were grouped into four factors—namely, family, spousal, social and professional support [73]. The FSS [72] was highlighted as a measure suitable for use with caregivers of children with disabilities in a scoping review on the subject [40], and was reported to have both internal consistency and stability across samples [73].

The participation of children was reported using the Children's Assessment of Participation and Enjoyment (CAPE) [71]. The CAPE [71] is a self-report questionnaire that has been developed for use with children/youth between the ages of 6 and 21 years, with and without disabilities. The CAPE [71] considers 55 activities grouped into domains (overall, informal and formal) or activity types (recreational, active-physical, social, skills-based and self-improvement). For each activity, five dimensions of participation are measured—namely, diversity, intensity, companionship, location and enjoyment [71]. A proxy report of the CAPE [71] was used to measure the caregivers' perceptions of their children's participation [74]. As reported in Dada, Bastable, Schlebusch and Halder [74] and available in this special edition of the Int. J. Eviron. Res. Public Health, the internal consistency of the CAPE [71] for this study was excellent (0.923 < α < 0.993) [74,75].

All materials for this study were translated into Afrikaans, Bengali, Sepedi, isiXhosa, and isiZulu. Translation included forward and blind backward translation, as well as the consideration of linguistic, functional and cultural equivalence [76].

#### *3.5. Data Collection*

A total of 422 information packs were sent to caregivers via their child's school. The return rate in India was approximately 70% and South Africa approximately 55%.

The information pack included information on the study, a written consent form, the FSS [72], and the proxy version of the CAPE [71] in the language of teaching and learning at the child's school. The consenting caregivers completed these forms and returned them to the school in an envelope. The children whose caregivers consented were asked to provide assent and they completed the CAPE [71] in an interview at their school. All children in India assented and 98% of South African children assented. The CAPE [71] interview was conducted in close adherence to the instructions and using the visual supports provided in the manual. The interview was conducted in the language of teaching and learning at the school or the child's home language, with the researcher reading the questions to the child and recording their answers on the CAPE [71] forms. Children who assented, as well as those who did not, were provided with a token of appreciation (a ruler and an eraser).

#### *3.6. Data Analysis*

Data analysis for this study was conducted using SPSS version 26 (IBM, Armonk, NY, USA) [77]. Demographic data, data from the FSS [72] and from the CAPE [71] were analysed for normality first, and then using non-parametric tests including Pearson's chi-square, Fisher's exact test, Mann–Whitney U, Kruskal–Wallace, and the independent samples test. Internal consistency of the FSS [72] was evaluated using Cronbach's alpha [75]. Due to significant differences being evident in the demographic data of participants from India and South Africa, the analysis of the CAPE [71] and associations between the CAPE [71] and FSS [72] were conducted on each set of data independently, rather than as a single set.

The participation data from India and South Africa were compared to their respective FSS [72] data using Kendall's Taub [78] to determine association. Although both Pearson's and Spearman's correlation coefficients are better known statistical coefficients, Kendall's Taub has been shown to be less sensitive to outliers, thereby limiting the number of Type 1 errors and providing tighter confidence intervals and clearer interpretation [78,79]—specifically where sample sizes are smaller [80].

#### **4. Results**

The internal consistency of the FSS [72] is reported on first, followed by the social support perceived by caregivers. This is followed by the associations in demographic and FSS [72] data. The participation data are summarised (the full data are available in the paper titled: The participation of children with intellectual disabilities: Including the voices of children and their caregivers in India and South Africa, in this special edition), and associations between social support and participation are presented.

#### *4.1. Internal Consistency of the FSS [72]*

The internal consistency of the FSS [72] was determined using Cronbach's alpha. The FSS [72] presented with Cronbach's alpha coefficients between 0.748 and 0.780, which are considered acceptable [75,81].

#### *4.2. Social Support Reported by Caregivers of Children with Intellectual Disabilities in India and South Africa*

On average, caregivers in India and South Africa reported that family and spouse groups were generally helpful (family mean = 2.72, spouse mean = 2.70), and social and professional groups were sometimes helpful (social mean = 1.67, professional mean = 2.23). The caregiver's parents and spouse were most likely to be reported as extremely helpful. Parent groups, co-workers, social groups, church or spiritual support, early childhood intervention centres, and governmental and non-governmental agencies were most likely to be unavailable to caregivers. No significant differences were evident between social support factors for India and South Africa, except for spousal support (*p* = 0.000). For specific support sources, significant differences were evident between India and South Africa, both for the level of support reported and the sources available. Overall, the caregivers in India reported greater helpfulness from available support sources, but older children, co-workers or parent groups, social and religious groups, early childhood intervention and governmental/non-governmental services were not available to the majority of families. The South African caregivers, however, reported that social support groups were less helpful to them. Unavailable support in South Africa included relatives, spousal friends, friends, neighbours, other parents, parent and social groups, early childhood intervention and governmental/non-governmental services. The full social support results are indicated in Table 2.




**Table 2.** *Cont*.

<sup>1</sup> Scores were measured on a Likert scale: 0 = not available; 1 = not at all helpful; 2 = sometimes helpful; 3 = generally helpful; 4 = very helpful; 5 = extremely helpful. <sup>2</sup> Pearson's chi-square, *p* < 0.05. <sup>3</sup> Mann–Whitney U, \* P is significant when *p* < 0.05.

#### Associations between Demographic Factors and Social Support

In India, the association indicated increased family support when the caregiver was the mother (Taub = 0.194), whereas in South Africa, decreased family support was indicated when the caregiver was the mother (Taub = −0.163). Small positive associations between social support (Taub = 0.201) and employment (Taub = 0.157) were evident in India, while a small effect of home language on professional support was indicated for South Africa (*p* = 0.262). Associations between child sex and age were seen in India, with increased support for male (Taub = −0.182) and younger children (Taub = 0.173) [79]. The association data are presented in Table 3 below.


**Table 3.** Associations between demographic social support factors in India and South Africa.

\* *p* < 0.05. <sup>1</sup> Taub. <sup>2</sup> Kruskal–Wallis.

#### *4.3. Participation and Social Support for Children with Intellectual Disabilities in India and South Africa*

#### 4.3.1. Self-Reported Participation of Children with Intellectual Disabilities

Children in India and South Africa participated in a similar number of activities overall and with the same enjoyment. However, children from India were noted to participate more frequently at home with close family, while the children from South Africa participated less frequently at a relative's house with extended family (medium to large effects). The full participation data are available in Dada, Bastable, Schlebusch and Halder, in this special edition of the Int. J. Eviron. Res. Public Health.

#### 4.3.2. Proxy-Reported Participation of Children with Intellectual Disabilities

Caregiver-reported participation differed from self-reported participation across both India and South Africa in terms of the number of social and skills-based activities participated in and to the level of enjoyment. No significant differences in the reporting of intensity, with whom, or where activities were conducted were evident. The full caregiver participation results are available in the Dada, Bastable, Schlebusch and Halder, in this special edition of the Int. J. Eviron. Res. Public Health.

#### *4.4. Association between Social Support and Participation*

4.4.1. Association between Social Support and Caregiver-Reported Participation of Children with Intellectual Disabilities in India and South Africa

Associations with the presence of family support sources were evident for the intensity of participation overall, in the informal domain and for self-improvement activities in South Africa. Family support was also associated with where formal and skills-based activities occurred in South Africa, but with caregiver-reported enjoyment in active-physical activities in India. Spousal factors were associated with the diversity of social activities in South Africa, with whom and where participation occurred overall, and with participation in the informal domain. In India, spousal support was associated with whom recreational activities occurred, and where informal and social activities took place. The social support factor was associated with the diversity of participation overall, in both the informal and formal domains in India. In South Africa, however, social support was associated with the intensity of participation overall, participation in the informal domain, and social activities. Intensity of participation in the formal domain as well as where participation in this domain occurred was associated with social support in India. Professional support was associated in India with the intensity of activities in the formal domain and social activities, but in South Africa, it was associated with participation with whom and participation in the informal domain. The association data are presented in Table 4.


**Table 4.** Association between social support factors and caregiver-reported participation.


**Table 4.** *Cont*.

Taub \*\* *p* < 0.01, \* *p* < 0.05. Effect sizes: <sup>1</sup> Small effect: Taub > 0.7, <sup>2</sup> Medium effect: Taub > 0.21, <sup>3</sup> Large effect: Taub > 0.50 [79].

4.4.2. Associations between Social Support and Child-Reported Participation of Children with Intellectual Disabilities in India and South Africa

Children in South Africa indicated associations between participation and family and social support, while children in India indicated connections between active physical activities and family. Intensity of participation in social (India) and recreational (South Africa) activities was associated with social support. An association between with whom recreational activities occurred and family support was evident in South Africa, while family support was associated with where participation occurred in India as well as South Africa. All effects identified were small [75]. Significant associations are reported in Table 5.

**Table 5.** Significant associations between social support factors and child-reported participation.



**Table 5.** *Cont*.

Taub \*\* *p* < 0.01, \* *p* < 0.05. Effect sizes: <sup>1</sup> Small effect: Taub > 0.7.

#### **5. Discussion**

Intellectual disability is one of the leading developmental disabilities in low- and middle-income countries [69,70]. For caregivers, a child with an intellectual disability can increase the stress and demands of parenting. Caregivers may find themselves solely responsible for ensuring that their child's rights are recognised and their needs are met [34–36]. Yet, in the face of increased demands, caregivers have reported a lack of support from outside of their immediate family [82,83]. Increased stress for caregivers can limit their ability to support their children and provide them with the required developmental opportunities [37,38] through participation in activities [6]. The relationship between the caregiver and the child's participation has been described as a related factor that may influence participation (the environment) [7–9,84]. For caregivers, however, social support has been described as a buffer to stress [37,38], which may increase their capacity to facilitate their child's participation. This study sought to describe the different types of social support experienced by caregivers of children with intellectual disabilities in India and South Africa (middle-income countries) and to identify whether there is an association between the social support reported by caregivers and the children's participation.

As discussed previously, the bulk of research on participation originated in high-income countries [22]. Hence, this study sought to provide information on participation of children with intellectual disabilities in two middle-income countries. In the initial analysis of demographic information from the participants it became clear that although both India and South Africa are middle-income countries, significant differences were evident in the demographics of the caregiver groups from these two countries. Education, income and employment differed significantly among the caregivers, with the South African caregivers reporting lower levels of education, income and employment. Such differences highlight the need for research across both low- and middle-income countries, as demographic differences alone make generalisation from one country to another challenging.

The presence of multigenerational households has been hypothesised to affect social support structures and to be widely prevalent in collectivist cultures. Nonetheless, only half of the families in this study came from multigenerational households, with Indian caregivers reporting significantly more multigenerational households than caregivers in South Africa—despite the fact that it has been suggested that multigenerational households may provide additional support for caregivers of children with disabilities [50–52]. Our study suggests that it cannot be assumed that households from traditionally collectivist countries will contain multiple generations, even if this has been a cultural norm in the past. Nowadays, industrialisation and globalisation have a clear impact on societal functioning [53].

In spite of the demographic differences identified between India and South Africa, social support from family, social and professional factors was reported as similar, but spousal support was significantly different. Caregivers from India reported more support from spousal sources—including their spouse, spouse's parents—and friends than did caregivers from South Africa, while approximately a quarter of South African caregivers reported that their spouse and relatives were not available. The lack of spousal availability in South Africa may result from the country's past, as the systematisation of migrant labour under apartheid split families by allowing only the working individual to stay in an urban area. As a result, families were divided, with mothers and children living in a different place from fathers. The forced separation of families under apartheid has had a significant effect on family structure in South Africa, which is still experienced today [85]. In Indian culture (in contrast), once married, some women would traditionally live with their husband's family, hence having a spouse available may also include the support of his family [53].

The demographic factors of relationship to child (family), employment (India, social), home language (South Africa, professional), child sex (India, family) and child age (India, social) were associated with social support reported by caregivers, although the associations showed small effects. The support experienced by caregivers may well be related to the social structure of the country, including how neighbours and friends support working parents, and cultural biases relating to sex and age [86–88]. For example, in India, male children are often revered while female children may be seen as a burden [89,90], while in South Africa professional support is most often available to caregivers in English or Afrikaans, which may not be their home language [91].

Overall participation of children with intellectual disabilities in India and South Africa was similar, but differences were evident in the formal domain, as well as in respect of active-physical and recreational activities. Weak positive associations with social support were evident across both the Indian and South African data in relation to the diversity of participation (mostly family support, but also spousal and social support). As participation in activities for children with intellectual disabilities requires (in many situations) the availability of the activities, as well as a partner to facilitate the child, the presence of additional family support may reduce the load on the caregiver and provide more opportunities for the child to participate. Similarly, spousal and social support may increase the number of opportunities for the child to participate.

The effect that social support given to caregivers of children with intellectual disabilities has on the participation of their child is evident in the associations identified between support sources and the caregiver-reported participation. In India, associations were seen most often between social and professional support and the formal domain, while in South Africa associations between spousal and social support were evident more often in the informal domain. These differences could be linked to the availability of resources. With lower income reported by caregivers in South Africa, it is possible that informal activities place less of a financial burden on caregivers. Importantly, however, the South African caregivers reported more households where the spouse was not available than did the Indian households. Thus, the association between both spousal and social support is logical, as when spousal support was not available, the South African caregivers relied on extended family and friends for support.

It is interesting to note the differences in associations evident between participation data as reported by the children and caregivers—this is in spite of the data not being significantly different [74]. The children's participation data were associated with family and social support, mostly in informal activities for South Africa, but primarily in formalised activities for India. Enjoyment was associated with professional support for active-physical activities in South Africa. This is a logical conclusion, in that children with special needs may require devices or support to participate in active-physical activities. Such support is often provided by professionals or professional organisations, for examples the special Olympics. Of interest, however, is that the enjoyment of formal (India) and self-improvement (South Africa) activities was associated with family support for children. In this regard it may be that self-improvement activities are participated in most frequently at home with family, or that these activities are most important to the family—hence all family members contribute towards the child's enjoyment in these areas.

Overall, the data from our study provide evidence that environmental factors play a role in the social support that caregivers of children with intellectual disabilities receive. Sequentially, social support plays a role in the participation of children with intellectual disabilities. Differences in the associations between social support and caregiver-reported participation point to demographic and cultural influences on the participation of children with intellectual disabilities. At the same time, differences in social support and participation associations between the child- and caregiver-reported participation data emphasise the subjective nature of social support and participation. Hence, results should not be generalised from one country to the next, even when aspects of their cultures appear similar at face value. When considering the participation of children with intellectual disabilities, the family environment should be examined as a whole, with reporting from multiple members in order to understand the factors that affect participation.

Considered in relation to current models of participation, the effects of social support were mostly weak, yet consistent across multiple areas of participation. Hence, they cannot be ignored in the consideration of participation of children with intellectual disabilities. Although current models of participation such as the fPRC [84] now include the child's context and environment, they have until recently focused more on the direct associations between the child and the environment.

#### *5.1. Recommendations*

Recommendations arising from this study include the exploration of the role of environmental factors in the participation of children with intellectual disabilities in other countries. Specifically, further research is recommended on the effects that social support interventions have on the participation of children who are typically developing and those with disabilities.

#### *5.2. Limitations*

The FSS [72] used in this study focuses on the perceived helpfulness of different types of social support but does not provide an opportunity for caregivers to report on the context of the support or to identify alternative supports that are needed [40]. Perhaps additional measures of the type of social support that is needed could have been included.

#### **6. Conclusions**

The social support provided to caregivers of children with intellectual disabilities in India and South Africa was similar in many respects. However, social support is sensitive to demographic factors such as employment and the relationship of the caregiver to the child. Caregivers of children with intellectual disabilities overwhelmingly reported a lack of social and professional support. In both India and South Africa, studies showed positive associations between participation and social support. For India, increased social support was associated with increased diversity of participation, while in South Africa it was associated with increased intensity of participation. Differences in results from different countries may preclude the generalisation of results relating to both social support and participation.

**Author Contributions:** The authors of this article contributed in the following areas: conceptualisation: S.D. and S.H.; methodology: S.D. and S.H.; validation: K.B. and S.D.; formal analysis: K.B.; investigation: S.D. and S.H.; resources: S.D. and S.H.; data curation: K.B., S.D. and S.H.; writing—original draft preparation: K.B. and S.D.; writing—review and editing: K.B., S.D. and S.H.; visualisation: K.B. and S.D.; supervision: K.B. and S.D.; project administration: S.D. and S.H.; funding acquisition: S.D. and S.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** Funding from the National Institute of Humanities and Social Sciences/Indian Council of Social Science Research is hereby acknowledged. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the funders.

**Acknowledgments:** The authors would like to acknowledge the postgraduate students from the Master's in the Early Childhood Intervention programme who assisted with data collection.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analysis or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Participation Restrictions among Children and Young Adults with Acquired Brain Injury in a Pediatric Outpatient Rehabilitation Cohort: The Patients' and Parents' Perspective**

**Florian Allonsius 1,\* , Arend de Kloet 1, Gary Bedell <sup>2</sup> , Frederike van Markus-Doornbosch 1, Stefanie Rosema 3, Jorit Meesters 1,4,5, Thea Vliet Vlieland 1,4 and Menno van der Holst 1,4,\***


**Abstract:** Improving participation is an important aim in outpatient rehabilitation treatment. Knowledge regarding participation restrictions in children and young adults with acquired brain injury (ABI) is scarce and little is known regarding the differences in perspectives between patients and parents in the outpatient rehabilitation setting. The aims are to describe participation restrictions among children/young adults (5–24 years) with ABI and investigating differences between patients' and parents' perspectives. At admission in 10 rehabilitation centers, patients and parents were asked to complete the Child and Adolescent Scale of Participation (CASP; score 0–100; lower score = more restrictions) and injury/patient/family-related questions. CASP scores were categorized (full/somewhat-limited/limited/very-limited participation). Patient/parent-reported outcomes were compared using the Wilcoxon signed-rank test. 223 patients and 245 parents participated (209 paired-samples). Median patients' age was 14 years (IQR; 11–16), 135 were female (52%), 195 had traumatic brain injury (75%). The median CASP score reported by patients was 82.5 (IQR: 67.5–90) and by parents 91.3 (IQR: 80.0–97.5) (difference = *p* < 0.05). The score of 58 patients (26%) and 25 parents (10%) was classified as 'very-limited'. Twenty-six percent of children and young adults referred for rehabilitation after ABI had "very-limited" participation. Overall, parents rated their child's participation better than patients themselves. Quantifying participation restrictions after ABI and considering both perspectives is important for outpatient rehabilitation treatment.

**Keywords:** participation; rehabilitation; acquired brain injury; pediatric; patient-report; parent-report

#### **1. Introduction**

Acquired brain injury (ABI) refers to irreversible damage to the brain which either has a traumatic cause; i.e., caused by external trauma (TBI) or a non-traumatic cause (nTBI); i.e., by internal causes [1]. It is a common diagnosis in children and young adults. The estimated yearly incidence rates in the Netherlands per 100,000 children and young adults are 288.9 (0–14 years) and 296.6 (15–24 years) for TBI and 108.8 (0–14 years) and 81.5 (15–24 years) for nTBI, respectively [2]. Due to natural brain adaptation, the majority of children and young adults with ABI will recover within the first year after brain injury [3]. However, on average, approximately 30% have persisting problems, and this group may benefit

**Citation:** Allonsius, F.; de Kloet, A.; Bedell, G.; van Markus-Doornbosch, F.; Rosema, S.; Meesters, J.; Vliet Vlieland, T.; van der Holst, M. Participation Restrictions among Children and Young Adults with Acquired Brain Injury in a Pediatric Outpatient Rehabilitation Cohort: The Patients' and Parents' Perspective. *Int. J. Environ. Res. Public Health* **2021**, *18*, 1625. https://doi. org/10.3390/ijerph18041625

Academic Editor: Dana Anaby Received: 30 December 2020 Accepted: 3 February 2021 Published: 8 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

from rehabilitation treatment [1–5]. One of the ultimate goals of (outpatient) rehabilitation treatment is optimizing a patient's daily life participation [2,6–10]. However, despite its relevance, knowledge on participation restrictions of children and young adults with ABI referred for rehabilitation treatment is scarce. The currently available literature focuses on children (<14 years) with TBI in hospital-based cohorts [10–18].

Only a few studies focus on both patients' and parents' perspectives, and knowledge regarding outcomes on participation measuring both perspectives is even more scarce [9,12,14,19,20]. Moreover, for the pediatric rehabilitation-based population, and in the context of family-centered care, the question is whether the severity and nature of participation restrictions can best be rated by patients, parents or both, which is still an under-researched area [20–24].

Two relevant studies (a study in the United States (US) and a Dutch study) found strong internal structure validity and internal consistency between the patient and parent reported versions of the outcome measures i.e., the Child and Adolescent Scale of Participation (CASP) [9,20]. Yet, discrepancies between patients' and parents' perspectives were found, where parents reported lower scores than the patients [9,20]. However, the study conducted in the US only focused on youth aged 11–17 years and with chronic conditions/disabilities, and making comparison to patients with ABI difficult [20]. The Dutch study focused on patients with ABI a small age range (14–25 years), and used a relatively small sample size (*n* = 49) from only one rehabilitation center [9]. This rehabilitation-based study in which the primary focus was on fatigue outcomes, investigated participation as well and found multidirectional relationships between participation and fatigue as well as considerable participation restrictions among patients with ABI as measured with the CASP (median 82.5, IQR 68.8, 92.3) [9].

Other studies based on hospital-based cohorts, report that 25–80% of children and young adults with either TBI (mild/moderate/severe) or nTBI (i.e., stroke, tumor) experience participation restrictions after ABI [2,6,7,9,10,14,16,17,25–36]. This wide range is due to differences in definition of participation, outcome measures, inclusion criteria (i.e., age, type and severity, hospital based) and time points (i.e., time since onset of ABI) used in these studies [36]. In both children and young adults, participation restrictions after ABI tend to persist for a long time which negatively influences life development [37]. Negative consequences could affect the development of physical, psychological and social emotional skills and competencies, as well as the shaping of identity, health and wellbeing in adulthood [2,7,9,16,17,25,30–36,38–40].

Regarding the factors associated with participation restrictions, several studies found that more participation restrictions after pediatric ABI were associated with (among others), diminished health-related quality of life (HRQoL), and negative patient and environmental influences i.e., more patient's motor, cognitive, behavioral and emotional consequences [7,12,16,22,23,36,41,42]. To date, these influences were not investigated among children and young adults with ABI who were referred for outpatient rehabilitation treatment.

The present study aims to investigate among children and young adults with ABI (5–24 years with TBI or nTBI) who were referred for outpatient rehabilitation treatment (not having received any prior rehabilitation treatment):


#### **2. Materials and Methods**

#### *2.1. Design*

Data from patients with ABI (and/or their parents) that were referred for outpatient rehabilitation treatment on the basis of continuing and/or expected problems, related to their brain injury were analyzed. These patients had not received any outpatient rehabilitation treatment yet. This study was part of a larger multi-center study on family impact, fatigue, participation and quality of life and associated factors in the Dutch ABI population (children and young adults). The study was started in 2015 in 10 Dutch rehabilitation centers, using a consensus-based set of patient/parent-reported outcome measures (PROMs) at admission as part of routine care. The reports of these PROMs were used for clinical goal setting in rehabilitation practice. The protocol for this study was reviewed by the medical ethics committee of the Leiden University Medical Center (P15.165), and an exempt from full medical ethical review was provided. For the current article the 'Strengthening the Reporting of Observational studies in Epidemiology' (STROBE) guidelines were used [43].

#### *2.2. Patients*

All children and young adults aged 5–24 years with a diagnosis of ABI, who were referred for outpatient rehabilitation treatment to a participating rehabilitation center and their parent(s) were eligible to participate. If patients and/or parents were unable/limited to write and/or understand the Dutch language, they were not invited by the center's health care professionals to complete the questionnaires. Patients over the age of 16 years had to give their parents' permission for completing the questionnaires according to the Dutch law of healthcare decision making.

#### *2.3. Data Collection*

Demographic and injury characteristics were extracted from the medical records by health professionals employed by the rehabilitation centers where patients had their appointment. For the outcomes related to participation, quality of life and child and environmental outcomes a (digital) questionnaire was administered to patients and/or their parents. Patients and parents were given the opportunity to complete this questionnaire prior to the first appointment during their visit at the outpatient rehabilitation clinic. If a patient (in case of a young adult) came without parents to the appointment, parents were asked to complete the questionnaires either on paper or digitally within one week after the first appointment. Unique links to the digital questionnaires were sent to the participants by e-mail by the medical health professionals working at the rehabilitation centers. Data were recoded, and thereafter anonymously stored in a central database at Basalt rehabilitation center in The Hague (The Netherlands). Finally, after analyzing the data, the centers received the results to use for clinical practice.

#### *2.4. Assessments*

#### 2.4.1. Demographic and Injury Characteristics

Information regarding demographics and injury-related characteristics included: date of birth, date of injury, date of referral to rehabilitation, age at the start of the first appointment i.e., the difference between date of birth and date of referral to rehabilitation and gender i.e., male/female. Time between onset of ABI and referral to rehabilitation was calculated and thereafter divided into 2 groups: referred for rehabilitation within 6 months, and after 6 months after ABI onset. The categorization of ABI was divided in: TBI/nTBI. If known, the TBI severity levels were divided into either mild, or moderate/severe (based on the Glasgow Coma Scale at hospital admission [44]). NTBI causes were divided into stroke/cerebrovascular accidents, brain tumors, meningitis/encephalitis, hypoxia/intoxication, and other.

#### 2.4.2. Participation Outcome Measure

The Child and Adolescent Scale of Participation (CASP) was administered to patients and parents to measure participation restrictions of the patient. The CASP is part of the "Child and Family Follow-up Survey" (CFFS) [45]. The CFFS, including CASP was validated for children, young adults and youth with ABI, was translated in the Dutch language, and is considered feasible and reliable tools to assess participation restrictions [2,17,20,25,45–48]. Patient-report (both children and young adults) and parent-report

versions of the CASP were available and used both in the present study [17,20,47]. The CASP is a 20-item questionnaire, yielding a total score, and 4 domain scores including: home & community living activities; 5 items, home participation; 6 items, community participation; 4 items, and school/work participation; 5 items. Activities regarding participation are rated on a 4-point scale: 4 = age expected (full participation), 3 = somewhat limited, 2 = very limited, and 1 = unable. Items marked as" not applicable" do not receive a score. Scores for each item are summed and divided by the maximum possible score based on the number of items rated. The results, multiplied by 100, give a final score between 0–100, which counts for both the total score and the domain scores. The higher the scores, the closer a patient is participating to age-expected participation levels in daily life.

#### 2.4.3. Four-Level Categorization

For the present study, a 4-level categorization system was developed to distinguish between levels of participation restrictions of patients for use in clinical practice. First, a draft version of a 4-level categorization was created by five of the authors based on preliminary analysis of the CASP data gathered for the present study and consensus discussions (F.A., A.d.K., M.H., G.B. and T.V.V.). We thereafter presented the categorization to a group of physicians and psychologists in the field, and to the remaining authors who are all experts in the field. Together, consensus was reached on the categorization and it was agreed to use it for further analyses in the present study. The 4-level categorization was made as follows:


#### 2.4.4. Secondary Outcome Measures

When assessing participation restrictions, patient (i.e., children and young adults) factors, environmental factors as well as health related quality of life were described using the following outcome measures:


for the for the pediatric TBI population) [50] It yields a total-score and 4 dimensionscores i.e., physical functioning (8 items), emotional functioning (5 items), social functioning (5 items), school/work functioning (5 items) [49] Items are answered on a Likert-scale (0 = never to 4 = almost always) and thereafter linearly transformed to a 0–100 scale (0 = 100, 1 = 75, 2 = 50, 3 = 25, 4 = 0). The results, items summed and divide by the number of items answered gives a final score between 0–100, with lower scores indicating diminished HRQoL [49,51].

#### *2.5. Statistical Analysis*

#### 2.5.1. Characteristics

Patients' injury, demographic and family related characteristics were described using descriptive statistics. All continuous variables were expressed as medians with interquartile ranges (IQR) and means with standard deviations (SD), based on their distributions (Kolmogorov-Smirnoff (K-S) test). Characteristics were presented for the total group and for the group of children (5–17 years) and the group of young adults (18–24 years) separately. The age categorization for children and young adults is in line with the Committee on Improving the Health, Safety, and Well-Being of Young Adults (Washington DC, 2015) and previous Dutch studies in patients with ABI [50,52–54].

#### 2.5.2. Primary/Secondary Outcome Measures

Regarding the primary (CASP) and secondary outcome measures (CAFI, CASE, PedsQL™ GCS-4.0), descriptive statistics were used to describe both the patient-report and the parent-report total scores of the CASP and the PedsQL™ GCS-4.0 and, if applicable, the domain scores. The CAFI and CASE were described similar as the CASP and the PedsQL™ GCS-4.0 but were only parent-report outcome measures. All outcomes were expressed as medians with IQRs (K-S test). To assess the potential correlation between the total scores of the CASP, PedsQL™ GCS-4.0 for HRQoL (patient/parent-report) and the CAFI/CASE (parent-report), Spearman correlations were calculated (Rho; ρ) and were considered: very strong, if >0.70; strong, if 0.40–0.69; moderate, if 0.30–0.39; weak, if 0.2–0.29; and negligible, if <0.19 [55].

#### 2.5.3. Four-Level Group Categorization (CASP)

To interpret how limited the patients' participation restrictions were (patient-report and parent-report), the 4-level group categorization was used i.e., "full participation"/ "somewhat limited"/"limited"/"very limited" participation. The CASP median (IQR) total scores are presented for all 4 group category levels. Per group (1 to 4), patient characteristics i.e., age, gender, time between administration to rehabilitation and ABI onset (<6 months or ≥6 months between onset and referral), cause; TBI/nTBI; and severity levels TBI; mild/moderate-severe, were reported (using descriptive statistics). Finally, within-group median (IQR) total scores of the CAFI/CASE/PedsQL™ GCS-4.0 were reported.

#### 2.5.4. Comparing Patients' and Parents' Perspectives

To compare outcomes, data from the patient-report and parent-report CASP versions, Wilcoxon signed-rank tests were used, for children and young adults separately. To test agreement between patients and parents additionally the Intraclass Correlation Coefficients (absolute agreement, single measures; ICC's) were calculated both for the CASP total and CASP domain scores. ICC scores were considered poor, if <0.40; moderate, if 0.41–0.60; good, if 0.61–0.80; excellent, if >0.81 [56]. Regarding the results obtained by using the 4-level categorization system, Weighted kappa (Kw) with linear weights was used to assess agreement between patients' and parents' scores [57,58]. The Strength of agreement is considered: poor, if < 0.20; fair, if 0.21–0.40; moderate, if 0.41–0.60; good, if 0.61–0.80; very good, if 0.81–1.00 [57–59]. A Bonferroni correction was performed to account for multiple testing (the α-value divided by the number of analyses on the dependent variable did not exceed 0.05). Outcomes were described for the total group, for children (5–17 years), and

for young adults (18–24 years) separately. Descriptive statistics were used to describe the CASP median (IQR) total scores, domain scores and categorization (counts, percentages). Differences/similarities in participation restriction categorization were described as follows: patients scoring in the same category as their parents, patients scoring themselves 1 to 3 categories lower than their parents, and patients scoring themselves 1 to 2 categories higher than their parents.

All analyses were performed using SPSS 24.0 for Windows (IBM, SPSS Statistics for Windows, Version 24.0. IBM Corp, Armonk, NY, USA). The level of significance was set at *p* < 0.05 for the Spearman Rho correlation, Wilcoxon signed rank and ICC tests.

#### **3. Results**

#### *3.1. Characteristics*

Patient, family and injury related characteristics are described in Table 1. The flow of all eligible participants for the current analyses can be found in Figure 1. The data of twohundred-sixty patients, (217 children (83%) and 43 young adults (17%)) and/or their parents was analysed. In total, there were 223 patient- and 245 parent-reported questionnaires completed and there were 209 patient-parent pairs (see Table 1 and Figure 1). One hundred and ninety-five (75%) patients had TBI of which 151 were mild TBI (77%). One hundred and thirty-five patients were female (52%). Ninety-six patients (39%) were referred to the rehabilitation center more than six months after brain injury onset. The median age of the patients in the group of children (5–17 years) was 14 years (IQR 11–16), and 18 (IQR 18–19) in the ≥18-year-old age group.

**Table 1.** Patient, family and injury characteristics of children and young adults with acquired brain injury (ABI) referred to an outpatient rehabilitation center.


\* Based on Glasgow Coma Scale at hospital admission: "mild"—13–15, "moderate"—9–12, "severe" < 8 \*\* Educational level parent: low prevocational practical education or less, intermediate—prevocational theoretical education and upper secondary vocational education, high—secondary education, higher education and/or university level education.


**Figure 1.** Flow of children and young adults with ABI admitted for rehabilitation and eligible for the present analysis. \* Missing participants: *n* = 11 no official ABI diagnosis, *n* = 12 incomplete questionnaires. # Number of filled out questionnaires used in this analysis (total/patient-reported/parent-reported): 1; number of questionnaires filled out by the patient, the parents or both in total and per age group (children, adolescents and young adults). 2; number of questionnaires filled out by parents only in total and per age group (children, adolescents and young adults). 3; number of questionnaires filled out by patients only (self-reported) in total and per age group (children, adolescents and young adults).

#### *3.2. Participation Outcomes*

Regarding participation outcomes in our population, as seen in Table 2, the median CASP total score reported by patients (*n* = 223) was 82.5 (IQR: 67.5–90.0), and by parents (*n* = 245) was 91.3 (IQR: 80.0–97.5). As seen in Table 2, Figure 2a,b, the lowest scores were found in the domain score "community participation" i.e., median patient-report score 75.0 (IQR: 56–92), median parent-report score 87.5 (IQR: 75–100). The highest median scores were found in the 'home participation' domain score for patients (87.5, IQR: 75–96), and in the "school/work participation" domain score for parents (95.0, IQR: 83–100).


**Table 2.** Total and domain scores on the CASP, CAFI, CASE and PedsQL™ GCS-4.0 (HRQoL) of children and young adults with acquired brain injury (ABI) and mutual correlations.

<sup>1</sup> CASP: Child and Adolescent Scale of Participation, 0–100 with lower scores indicating more participation restrictions. <sup>2</sup> PedsQL™ Generic Core Scales 4.0 for Health-related quality of life (HRQoL): 0–100 with lower scores indicating lower HRQoL. <sup>3</sup> CAFI: Child and Adolescent Factors Inventory (CAFI), and CASE: Child and Adolescent Scale of Environment, 0–100 with higher scores indicating more problems. \$ ρ = Spearman's rho (ρ) correlation. \*\* *p* < 0.001.

Secondary outcome measures are also presented in Table 2. Regarding HRQoL, the median PedsQL™ GCS-4.0 patient-report total score was 65.2, (IQR: 53–78), and the median parent-report score was 60.9 (IQR: 48–75). The parent-report median scores in the CAFI (child/young adult factors) and CASE (environmental factors) were: 56.9 (IQR: 49–65) and 39.0 (IQR: 33–51), respectively. Spearman's rho correlations between the CASP scores and the CAFI/CASE and HRQoL were significant (*p* < 0.01) and strong ranging between: ρ 0.53–0.67.

#### *3.3. Four-Level CASP Categorization*

Table 3 shows within-group (patient/injury-related) characteristics, and CASP/CAFI/ CASE/HRQoL scores of participation restrictions (patient-report and parent-report where applicable) in our cohort, organized by the 4-level CASP participation restrictions categorization. Eighty-nine percent of the patients, and 73% of the parents reported patients' participation restrictions in more than one CASP domain. Forty-three percent (patientreported) and 45% (parent-reported) reported CASP total scores that fell in the "somewhat limited" category. Twenty six percent (patient-report) and 10% (parent-report) reported CASP total scores that fell in the "very limited" category. In this "very limited" category, median CASP scores were 57.9 (IQR: 50–64) for patient-report data, and 61.4 (IQR: 49–65) for parent-report data. Patients who fell in this 'very limited' category, had a median age of 15 years (both in the patient and parent-reported category), 45–52% were female, 64–78% had a TBI and 33–40% were referred for rehabilitation more than 6 months after ABI onset. Lower participation CASP scores, i.e., category levels up to category 4, also showed lower (diminished) patient and parent report HRQoL scores, and higher (more problems) parent report CAFI/CASE scores.

**Figure 2.** (**a**) Differences in CASP scores between Patients and Parents in children (5–17 years) with ABI. \* CASP: Child and Adolescent Scale of Participation, 0–100 with lower scores indicating more participation restrictions. (**b**) Differences in CASP scores between Patients and Parents in young adults (18–24 years) with ABI. \* CASP: Child and Adolescent Scale of Participation, 0–100 with lower scores indicating more participation restrictions.


 CASE: Child and Adolescent Scale of Environment,

 0–100 with higher scores indicating more problems.

b

Parent-report

 CAFI: Child and Adolescent Factors Inventory, and Parent-report

*Int. J. Environ. Res. Public Health* **2021**, *18*, 1625

#### *3.4. Differences in Patients' and Parents' Perspectives*

In Table 4, the differences in participation outcomes between patients and parents (paired samples) is reported. Regarding the total paired-sample group (*n* = 209), there was moderate agreement in participation total CASP and domain outcomes between patients and their parents i.e., ICC = 0.42–0.57, all *p* < 0.001. In the group of children (5–17 years, *n* = 176) moderate agreement was found between patients' and their parents' total CASP and domain scores (ICC = 0.43–0.55, all *p* < 0.001). In the young adult (≥ 18 years, *n* = 33) group, there was poor-moderate patient/parent agreement between patient- and parent report scores on all CASP domains (ICC = 0.37–0.59, all *p* < 0.001). Regarding the categorical data on the 4-level categorization system, a fair to moderate agreement was found between the patients and parents; "moderate" in children; Kw: 0.42 (95%CI 0.32–0.52, *p* < 0.001), and "fair" in young adults; Kw: 0.27 (95%CI 0.08–0.46, *p* < 0.05). Regarding the differences in categorization between patients and their parents, in the total paired-sample group, 38% of the patients scored themselves in a lower CASP level category than their parents. In the group of children, the same percentage was found (38%), while in the young adult group 51% scored themselves in a lower category than their parents.

**Table 4.** Differences and similarities between patient and parent CASP participation scores and categories.



**Table 4.** *Cont*.

<sup>1</sup> CASP: Child and Adolescent Scale of Participation, 0–100 with lower scores indicating more participation restrictions. # Z scores for Wilcoxon signed-rank test for nonparametric data outcomes \* *p* < 0.05, \*\* *p* < 0.001; \$ ICC; Intraclass Correlation Coefficients rated: <0.40: poor; 0.41–0.60: moderate; 0.61–0.80 good; >0.81: excellent. Kw: Weighted Kappa interpretation (categorical CASP score): <0.20: poor, 0.21–0.40: fair, 0.41–0.60: moderate, 0.61–0.80: good, 0.81–1.00: very good—agreement. ˆ Patient categorization compared to parents' categorization: The differences in categorized participation between patients and their parents, a: Patients that scored 1 category worse than their parents, b: Patients that scored 2 categories lower than their parents, c: Patients that scored 3 categories lower than their parents, d: Patients that scored 1 category better than their parents, e: Patients that scored 2 categories better than their parents.

#### **4. Discussion**

According to data gathered before/on the first appointment for routine outpatient rehabilitation for children and young adults with ABI and their parents in multiple rehabilitation centers, 88% (patient-reported) and 73% (parent-reported) of the patients have participation restrictions that can be classified as "somewhat limited" to "very limited", with a considerable number of patients (25, parent reported and 58, patient reported) that can be classified as "very limited". The large majority was classified in the "somewhat limited" category. Overall, patients consistently reported more severe participation restrictions than parents. There was a greater discrepancy in the levels of participation restrictions between patients and parents in the young adult group compared to the children group.

#### *4.1. Participation Restrictions*

These results confirm that experiencing participation restrictions is common in pediatric patients with brain injuries (TBI/nTBI) [2,6,7,9,12,16,17,25,30–36,41]. Furthermore, the results we found, pointed out that the rehabilitation referred group had more participation restrictions compared to a Dutch hospital-based cohort [2]. In the current analyses of data among patients referred to an outpatient rehabilitation center, the vast majority reported participation restrictions in one or more domains of the CASP. This proportion was relatively high as compared to the 25–80% reported in a systematic review of studies on participation restrictions in children and youth with ABI including in hospital-based cohorts [7]. The current analyses found that the majority of patients was classified as "somewhat limited". These patients could also be "at risk" regarding restricted participation. In clinical practice it could also be important to monitor the patients that score relatively better than patients with more limited participation. However, future research must confirm this hypothesis by further looking into the "somewhat limited" patients. Concerning the prevalence of participation restrictions in young adults, some differences with the literature were found. A previous rehabilitation-based study, with patient and parent-reported data that focused on patients with ABI in the age group of 14–25, reported similar participation restrictions when compared to the results of the total sample from the

current analyses [9]. However, more participation restrictions were found in the young adult group in the current analyses [9]. Differences could possibly be explained by differences in age inclusion. Results suggest that young adults experience more participation restrictions than children. This could be explained by the greater appeal made on for example independence, planning and coping in this transitional age group.

#### *4.2. Community Participation*

For both patient-report and parent-report CASP outcomes and in all (age) groups (<18 years/>18 years/total), the lowest scores were found in the domain 'community participation' which includes participation related to e.g., social play/leisure activities with friends, events, sports, doing groceries, communicating with others in the neighborhood. [45,47]. Restrictions in community participation could also be related to the fact that children and young adults with ABI often have difficulties in social functioning, emotional functioning, and processing sensory stimuli (after ABI onset). These competences are needed when participating in the community [7,37]. However, other factors (e.g., environmental resources, stigma, family support, as well as time allocation), may also influence community participation [14,42].

#### *4.3. Correlations with the CASP and CAFI/CASE/HRQoL*

In comparison to a previous Dutch study in a hospital cohort with a higher CASP total score, the mutual correlations of the CASP with the CAFI, and CASE (parent-report), were higher in this rehabilitation-based population [2]. Regarding HRQoL, in line with previous literature participation was found to be highly correlated with HRQoL (patient-report and parent-report) [9,16,35]. These results underline the interdependence of limitations on the level of participation (CASP), child/young adult factors e.g., body functions and structures (CAFI), environmental factors (CASE) and HRQoL (PedsQL GCS-4.0). These findings also support the assumption that the CASP, PedsQL GCS-4.0, CAFI and CASE are more suitable among patients that were administered to outpatient rehabilitation (and filled out the questionnaires at admission) than in patients that were in a hospital (hospital-based).

#### *4.4. Notable Results Found in the Current Rehabilitation-Based Population*

Notable results were found in the current analyses among the outpatient rehabilitationbased population, which were not found in previous studies [36,41].


These findings should be discussed with professionals in acute care to increase awareness of possible consequences of later rehabilitation referral and to ultimately improve referral policies and procedures.

#### *4.5. Differences in Perspectives*

Regarding patients' and parents' perspectives, moderate agreement between patient and parent reported CASP (total and domain) scores were found. Previous studies underlined the importance of measuring both patients' and parents' perspectives to assess outcomes [20,36]. One Dutch study regarding adolescents and young adults with ABI found a difference between the patients and the parents CASP total score outcomes, similar to what we found in the results of the analyses [9]. Parents tend to report less participation restrictions for their children than the patients themselves, which is in contrast to previous studies with other outcomes (e.g., HRQoL; where parents usually report lower scores than their children) [9,16,17,25,30–33,35,40]. This was also found in our analyses. A large part of the patients in our cohort scored themselves in another CASP level category than their parents did. These discrepancies in reporting outcomes may be explained due to the fact that most participation activities (of the children and young adults) occur outside of the home environment where parents are not present and also, young adults spend more time away from parents than children. Our results suggest that assessing both patients' and parents' perspectives is important in order to identify differences and similarities. By using both perspectives, a broader view on overall functioning is attained, providing health care professionals the opportunity to consider both patients' and parents' perspectives when collaborating on rehabilitation goals, and make sure parents play an active role in today's often proposed family-centered care [14].

#### *4.6. Categorization of Severity of Participation Restrictions*

In the currently analyzed data, a 4-level categorization was created that correspond to specific CASP score ranges to reflect the overall degree of participation restriction. This categorization was based on previously identified levels of participation suggested by one of the authors (G.B.). To date, CASP outcomes were described as just a score between 0 and 100 (lower score = more participation restrictions). To facilitate a better interpretation of the score in clinical practice, we proposed a categorization of the total score into four levels. This 4-level categorization can be used next to the original 0–100 score) to compare and report CASP outcomes. The use of cut-off values may help to contextualize and differentiate the scores for clinical practice (i.e., indication for rehabilitation, evaluation of intervention) and research. All statistical comparisons of patients' and parents' scores in the present study consistently demonstrated a considerable discrepancy. Poor agreement was also seen using the proposed 4-level categorization, substantiating the validity of that division. Regarding the 4 categories, the majority of the patients and their parents reported CASP scores in the 'somewhat limited', the 'limited' and 'very limited' categories. A quarter of the children and almost one-third of the young adults scored in the most restricted, i.e., "very limited" category. Parent and patient-report scores differed in participation restriction category in almost half of the of cases, with parent scores and categories demonstrating lower levels of participation restriction as previously described. Future longitudinal studies could use this new categorization to further evaluate its utilization, and/or to investigate recovery outcomes over time (e.g., moving to higher category level of participation) during rehabilitation treatment related to interventions.

#### *4.7. Limitations*

Describing analyses and results among rehabilitation referred patients resulted in a number of limitations. First, there was a relatively small sample of young adults compared to the sample of children (43 vs. 217). The explanation is merely organizational: most rehabilitation centers have a separate pediatric (<16 or <18 years) and adult (≥18 years) department where only the pediatric department was involved. Only two centers had a separate department for young adults (18–25 years) and included young adults. However, the number of included young adults was large enough to analyze and report outcomes for separately. Since, due to age and life phase, in the young adult group is a different group of patients it is recommended to include this group of patients in transition fully

in future pediatric studies. Secondly, not for all patients paired sample data was available, making the analysis for the differences/similarities between patients' and parents' perspectives only possible for a portion of our analyses (*n* = 209). However, since we had paired sample data available for the majority of patients, we believe that outcomes are generalizable. Third, the CASP is known to have a ceiling effect [17,47] Nonetheless, in contrast to other studies reporting ceiling effects in children and young adults with ABI, these were less evident in the present analyses making the CASP a more suitable instrument for use in rehabilitation cohorts (versus patients that are hospitalized) of patients with ABI [2,17,20,47]. Furthermore, an alternative instrument that also focusses on the ABI population is lacking [17]. Finally, results of patient/parent rated outcome measures could be biased, i.e., by limitations in motivation or patients' and/or parents' moment bound 'stress and mood'.

#### *4.8. Directions for Future Research*

Interesting follow up projects could be longitudinal studies monitoring participation over time and evaluative studies using the CASP to explore the effect of rehabilitation programs for children and young adults with ABI and their families, since optimizing participation is an important rehabilitation goal. In these studies, the newly developed categorization of participation outcomes could be used and further investigated on its usefulness and robustness. Future studies should include the search for the best available participation outcome measures particularly given the number of promising participation-focused, multi-setting interventions that recently have been developed to improve participation outcomes for individual children, youth, and families [21–24]. The next challenge is to drive implementation of participation-based interventions on a larger scale, and research should be focused on enabling strategies and on cost-effectiveness of these interventions. The CASP and our newly proposed categorization of participation restrictions could support this process.

#### **5. Conclusions**

A substantial portion of patients (ages 5–24 years) with acquired brain injury referred to an outpatient rehabilitation center in The Netherlands had "limited" to "very-limited" participation. Patients reported greater participation restrictions than their parents and disparities between patient reported and parent reported participation restrictions were greater in young adults than in children. Furthermore, a strong correlation was found between patient and environmental factors (CAFI and CASE), HRQoL (PedsQL GCS-4.0), and participation (CASP). Most restrictions were found in the 'community participation' domain. A large part of the patients with a late referral (>6 months) to rehabilitation after ABI onset reported "very limited" participation. Early referral is important as this may reduce participation restrictions. Taking into account both patients' as well as parents' perspectives is important in outpatient rehabilitation treatment in order to guide both patients and their parents appropriately during treatment. Furthermore, the categorization of CASP scores into 4 categories might be useful for clinical practice and research, but more study is needed to understand how this can be applied and inform participation focused clinical and practical decisions.

**Author Contributions:** All authors listed have contributed sufficiently to the project to be included as authors, and all those who are qualified to be authors are listed in the author byline. The authors that contributed in this research were: F.A., A.d.K., G.B., F.v.M.-D., S.R., J.M., T.V.V., M.v.d.H. and were responsible for: Conceptualization, F.A., A.d.K. and M.v.d.H.; methodology, F.A., M.v.d.H.; software, F.A., M.v.d.H.; validation, F.A., G.B., A.d.K., M.v.d.H.; formal analysis, F.A., M.v.d.H.; investigation, F.A., M.v.d.H., A.d.K.; resources, A.d.K.; data curation, F.A., A.d.K., M.v.d.H.; writing—original draft preparation, F.A.; writing—review and editing, A.d.K., G.B., F.v.M.-D., J.M., T.V.V., S.R., M.v.d.H.; visualization, F.A., M.v.d.H.; supervision, T.V.V., A.d.K., M.v.d.H.; project administration, A.d.K., T.V.V.; funding acquisition, A.d.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Hersenstichting (PZ2015.01.10) and the Revalidatiefonds (R2014.124).

**Institutional Review Board Statement:** The protocol for this study was reviewed by the medical ethics committee of the Leiden University Medical Center (P15.165), and an exempt from full medical ethical review was provided.

**Data Availability Statement:** Data used in this study is stored in a central database at Basalt Rehabilitation center, The Hague in the office of innovation, quality and research and can be available when requested.

**Acknowledgments:** We would like to thank all the patients and their families participating in this study. Further we would like to thank Cedric Kromme, and Åsa Mennema for contributing to this study by collecting and processing the data into the central database. Finally, we would like to thank all clinical health care professionals and medical secretaries of the participating rehabilitation centers contributing in this study i.e., Basalt in the Hague, De Hoogstraat in Utrecht, Heliomare in Wijk aan Zee, Vogellanden in Zwolle, Klimmendaal in Apeldoorn, Revalidatie Friesland in Beesterswaag, Libra Rehabilitation&Audiology in Tilburg, Revant Breda, Merem in Hilversum and Reade in Amsterdam.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Participation Profile of Children and Youth, Aged 6–14, with and without ADHD, and the Impact of Environmental Factors**

**Tair Shabat 1, Haya Fogel-Grinvald 1, Dana Anaby <sup>2</sup> and Anat Golos 1,\***


**Abstract:** Background: Children and youth with attention deficit hyperactivity disorder (ADHD) may experience difficulties in participation, but few studies examine their participation and the environmental factors affecting participation. This study explored the participation and the environmental factors of children and youth, with and without attention deficit hyperactivity disorder (ADHD), in the following three settings: home, school, and community. Materials and Methods: Parents of 65 participants aged 6–14 (M = 9.91, SD = 1.87) with and without ADHD completed the Participation and Environment Measure for Children and Youth (PEM-CY) questionnaire, which evaluates participation and environmental factors, along with demographic and screening questionnaires. Results: The ADHD group (*n* = 31) scored significantly lower than the non-ADHD group (*n* = 34) in "frequency" at home, "involvement", and overall environmental support in all settings, with parents expressing a greater desire to change their child's home and community participation. For the ADHD group, a relationship was found between environmental support and involvement in all three settings. Conclusions: The findings demonstrated differences in the participation of children and youth with ADHD across different settings, compared to those without ADHD, and confirmed the effect of environmental factors on participation, especially involvement. It is essential to consider participation measures and environmental factors when designing interventions for children and youth with ADHD.

**Keywords:** children and youth; ADHD; participation; frequency; involvement; environment; well-being

#### **1. Introduction**

The International Classification of Functioning, Disability and Health (ICF) of the World Health Organization (WHO) defines participation as "involvement in a life situation" [1], which is considered an important outcome measure for rehabilitation and intervention [1,2], as well as the focus and goal of many health and rehabilitation disciplines [3]. Participation in meaningful activities has a positive influence on health and well-being and is essential for the development of a person's abilities and self-efficacy [4,5], as well as for skill acquisition and learning among children and youth [6]. Participation is a multidimensional concept that includes various dimensions such as frequency and involvement. Participation frequency is considered as an objective aspect, referring, for example, to the number of times a person participates in an activity [7], while involvement refers to the feelings and personal experience of participation and includes various elements such as motivation, adherence, satisfaction, and emotional engagement [7–10].

The participation of children and youth with different health conditions (such as attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) and/or disabilities was found to be limited, compared to that of children with typical development [11–13]. For example, it was found that children with disabilities participate

**Citation:** Shabat, T.; Fogel-Grinvald, H.; Anaby, D.; Golos, A. Participation Profile of Children and Youth, Aged 6–14, with and without ADHD, and the Impact of Environmental Factors. *Int. J. Environ. Res. Public Health* **2021**, *18*, 537. https://doi.org/10.3390/ ijerph18020537

Received: 24 December 2020 Accepted: 7 January 2021 Published: 11 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

less frequently and/or are less involved in activities in the home, school, and community settings, compared to their peers with typical development [14–16].

One of the health conditions that may affect the participation of children and youth in different settings is attention deficit hyperactivity disorder (ADHD). This is a neurodevelopmental disorder characterized by attention deficit and/or impulsivity and hyperactivity, whose prevalence among children and youth is estimated at about 5%. Symptoms of ADHD persist in adulthood, with prevalence among adults estimated at about 2.5% [17,18]. A diagnosis of ADHD is based on the appearance of six or more criteria (such as lack of attention to details, difficulty organizing tasks, etc.), related to inattention, hyperactivity, and/or impulsivity, some of which appear prior to age 12; and these symptoms must occur in at least two life environments in a way that impairs functioning and quality of life [17].

ADHD has far-reaching and long-term consequences in all functioning areas, as it affects various aspects of a person's life, including daily functioning, employment, social participation, and family stability [17,18]. Studies among ADHD populations have often focused on specific impairments and/or functioning areas in which the implications of ADHD arise. For example, Shimoni et al. [19] and Engel-Yeger and Ziv-On [20] reported that children with ADHD participate less frequently in most leisure activities, receive less enjoyment from formal leisure activities, and show a lower preference for participation in some leisure activities, such as physical and social leisure activities, compared to children without ADHD. Additional studies indicated difficulties in social functioning of children and youth with ADHD compared to their peers [21,22], affecting their participation in various social activities [19,20].

Social difficulties of children and youth with ADHD may include peer rejection, inappropriate behavior, and/or difficulties with social skills such as collaboration, taking turns, reciprocity, and focus on conversation and play [21,22]. In addition, this population may experience difficulties in academic functioning [23,24], putting them at greater risk for low academic achievement, suspensions and expulsion from school, more absences, and even dropping out permanently, compared to populations without ADHD [25,26]. While these studies focus on one area of functioning, Harpin [27] described difficulties in the participation of a population with ADHD in multiple settings, and Lavi et al. [28] also reported significantly lower participation of adolescents with ADHD in daily activities and in school and home participation, compared to their peers without ADHD.

In summary, most studies examining the functioning of children and youth with ADHD are often focused on their specific difficulties and disabilities rather than their overall participation. In addition, there are few studies describing the participation profile of children and youth with ADHD compared to their peers without ADHD, particularly with respect to different settings (home, school, and community). Thus, a need exists to expand professional knowledge of the effects of ADHD on the participation of children and youth with this diagnosis, as it impacts their daily life in various settings [19].

In addition to health conditions, various environmental factors, such as physical and sociocultural factors, may also affect a person's development and participation [1,3,29,30]. Environmental factors can either support or hinder (erect barriers to) participation [3]. It is therefore important to examine the environmental factors and their impact on participation [31]. Not surprisingly, people with disabilities often identify relationships between environmental factors, participation, and quality of life. This highlights the need to assess the environmental impact on their participation at the community and social levels [29]. Research has found that the environmental domains noted in the ICF influence the participation of children with disabilities [32]. For example, the study by Bedell et al. [14], which examined community participation of children with and without disabilities, indicated that parents of children with disabilities more often rated environmental factors as barriers to participation and more rarely as supports, compared to parents of children without disabilities. Furthermore, it appears that the environment plays a unique role in influencing participation in different settings (home, school, and community). Specifically, environmental barriers were found to directly affect the frequency of

participation and involvement in all settings, whereas environmental supports only influenced involvement in home and community settings, and participation frequency in the community setting [11].

However, despite increasing research on the contribution of environmental factors in explaining the participation of children and youth with disabilities, the majority of these studies have focused on children with physical disabilities [32], autism [33], or developmental coordination disorder (DCD) [34]. A number of studies conducted among children and youth with ADHD have addressed the relationship between environment and functioning. They have identified factors such as attitudes and social–family environment as relevant in conducting a functional assessment of this population [35,36]. One qualitative study showed that over half of the participants described a particular aspect of ADHD as context-dependent, which may indicate an association between environment and ADHD symptoms [37]. In a review article, Dvorsky and Langberg [38] reported that social and family support, particularly social acceptance and positive parenting, has a positive effect and can even prevent the negative effects of ADHD. They also noted that research examining such supportive and protective factors is still in its infancy, suggesting further exploration.

All this points to the need for an in-depth examination of how environmental factors impact the participation of children and youth with disabilities [32]. Specifically needed is a comparison of those with ADHD to their peers without ADHD, in a range of settings, with attention given to the relationship between environmental supports and participation patterns. Our study addressed this need by examining the participation profile of children and youth with ADHD, and the environmental factors that may influence their participation. The results may contribute to a deeper understanding of the participation of children and youth with ADHD, thereby assisting in assessment and contributing to the development and implementation of appropriate intervention programs for this population.

The present study focuses on examining the participation of children and youth (aged 6–14) with and without ADHD, in home, school, and community settings, identifying the environmental supports for and barriers against participation, and examining the availability of supporting resources. Our specific objectives were to examine the following: (a) the differences between children and youth with and without ADHD with respect to participation patterns (in terms of frequency, involvement), desire for change, and the overall environmental support in each of the settings (home, school, and community); (b) the relationship between the overall environmental support and the participation patterns (frequency and involvement) in the different settings in each group (with and without ADHD); and (c) the differences in the participation patterns between the different settings (home, school, and community), and to describe the environmental factors (support and barriers) that influence participation of children and youth with ADHD.

#### **2. Materials and Methods**

#### *2.1. Study Design*

A descriptive quantitative and comparison cross-sectional design was used.

#### *2.2. Participants*

The study population included 65 parents of children and youth, aged 6–14, most of them (about 60%) from urban areas in the central district of Israel, who were recruited by voluntary response sampling. Most parents were married (89.5%), were aged 45 and under (mothers: 73.4%; fathers: 65.1%), and had an academic education (mothers: 83.6%; fathers: 74.6%). The participants were divided into the following two groups: (a) an ADHD group (31 participants) and (b) a non-ADHD group (34 participants) matched in age range and adjusted for gender and socioeconomic status (according to the level of family income). The inclusion criteria for each group were children and youth whose parents reported that they did (ADHD group) or did not (non-ADHD group) receive a diagnosis of ADHD from a qualified professional, with all reports confirmed by the ADHD Questionnaire (see

Instruments, Section 2.3). The exclusion criteria for both groups were as follows: (a) the parents' lack of fluency in the Hebrew language; (b) attendance of the child/adolescent in a special education environment; and (c) one or more of the following diagnoses for the child/adolescent: cerebral palsy, autism, epilepsy, Tourette syndrome, intellectual disability, psychiatric disorder, and/or brain injury, according to their parents' report in the demographic questionnaire (see Instruments, Section 2.3). In the ADHD group, at least half of the parents reported learning, sensory, and emotional–behavioral difficulties of their children. As seen in Table 1, both groups included participants with an average age of 9–10 years, most were boys (over 64%), and were from average or high socioeconomic strata. No significant differences were found in the characteristics between the two groups with respect to age, gender, or family socioeconomic status.


**Table 1.** Demographic characteristics of the participants.

Attention deficit hyperactivity disorder (ADHD); a—Independent-samples *t*-test; b—Chi-squared test for independence.

#### *2.3. Instruments*

#### 2.3.1. A demographic Questionnaire

A demographic questionnaire was developed for this study as a parental reporting tool. Its purpose was to characterize the study population, as well as to provide information about the child/adolescent and his/her family. The personal details in the questionnaire included items such as age, gender, country of birth, educational framework, health condition, and family income.

#### 2.3.2. The Attention Deficit and Hyperactivity Screening Questionnaire

The Attention Deficit and Hyperactivity Screening Questionnaire [17] is a parentreport questionnaire that identifies symptoms of ADHD according to the criteria found in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Version (DSM-5) [17]. It includes 18 criteria rated by the parent on a scale of 4 grades (3 = "very much", 0 = "not at all"). The questions are divided into criteria related to attention and hyperactivity, with suspected ADHD indicated by a score of 2 or higher in at least 6 of the 9 criteria for attention, and/or in at least 6 of 9 criteria related to hyperactivity and/or impulsivity. This questionnaire was used as an exclusion criterion for the non-ADHD group.

#### 2.3.3. The Participation and Environment Measure for Children and Youth (PEM-CY)

The Participation and Environment Measure for Children and Youth (PEM-CY) [8] is a parent-report instrument that examines participation and environmental factors affecting the participation of school-age children (5–17 years of age) across the following three settings: home, school, and community. The PEM-CY participation items represent broad types of activities typically performed in each setting, i.e., home (10 activities), school (5 activities), and community (10 activities). For each activity type, parents are asked to note the following: (a) how frequently their child participates ("never" = 0 to "daily" = 7); (b) how involved their child is while participating ("minimally" = 1 to "very" = 5); and (c) whether they desire change in their child's participation ("no" or "yes"; if "yes", parents identify the type of change desired: "frequency", "involvement", and/or "variety"). Parents are then asked whether certain features of the environment help or hinder their child's participation in activities in each setting ("not an issue", "usually helps", "sometimes helps/makes harder", "usually makes harder"). They are also asked about perceived adequacy/availability of supporting resources ("not needed", "usually yes", "sometimes yes/no", "usually no"). The PEM-CY has been found to have moderate-to-good internal consistency (Cronbach's α = 0.59–0.83) in the participation scales. Its test–retest reliability was found to be moderate at school (r = 0.58), and high at home (r = 0.84) and in the community (r = 0.79). Additionally, high reliability was found in the environment scale (r > 0.76). This measure identifies significant differences in participation patterns and environmental factors between children with and without disabilities [12], supporting its construct validity. It also has been effectively used in the Israeli context.

#### *2.4. Procedure*

This study was approved by the Ethics Committee of the Hebrew University, Jerusalem, Israel (No. 27032018). Ads for recruiting subjects were posted on social networks and relevant forums; for the ADHD group, therapists working with children and youth with ADHD were also contacted. Parents who showed interest and agreed to participate in the study received an explanatory letter and filled out the questionnaires electronically, via email, or manually, according to their preference. The data were collected without identifying personal and/or computer information. Screening of the returned questionnaires was performed according to the exclusion and inclusion criteria (see Participants, Section 2.2).

#### *2.5. Data Analysis*

Statistical analysis was performed using the SPSS version 25.0 (Statistical Package for the Social Sciences, Armonk, NY, USA) [39], with a significance level of 0.05. Descriptive statistics were used to describe the study population, including participants' background data distribution, and the environmental factors (supports and barriers) impacting the ADHD group. Differences between the two groups in gender and socioeconomic status were examined using the chi-squared test for independence, and an independent-samples *t*-test was used for the age variable. For each setting (home, school, community), the participation patterns (frequency and involvement) were measured using mean variables. The desire for change in their child's participation was measured as the percent of activities in which the parents indicated that desire. In addition, according to the PEM-CY manual's guidelines, a variable was calculated for the overall environmental factors supporting participation ("overall environmental support" (in each setting separately. All environment questions were recoded into 3-point scale by merging "Not an issue/Not needed" with "Usually helps/Usually yes". The sum of all the environment ratings was divided by the maximum possible score within a setting, and multiplied by 100 (higher scores indicated more support of children's participation or more availability of the supporting resource). In order to compare the participation patterns, the desire for change, and the overall environmental support between the two groups (with and without ADHD), one-way MANOVA analysis was used. Effect-size calculation was conducted using partial eta squared, with η<sup>2</sup> > 0.14 defined as high effect, 0.06 < η<sup>2</sup> < 0.13 as medium, and 0.01 < η<sup>2</sup> < 0.05 as low [40]. In order to examine the relationship between the overall environmental support and the participation patterns (frequency and involvement) within each group, Pearson's correlations were calculated. In order to examine the differences among the three settings in the participation patterns (frequency and involvement) of the ADHD group, a one-way repeated measures ANOVA was used with Bonferroni correction.

#### **3. Results**

*3.1. Comparison of Participation and the Overall Environmental Support between Groups in the Different Settings*

Differences in participation patterns between the two groups were examined using a one-way MANOVA analysis. The differences between the groups on the combined dependent variables were statistically significant for frequency (*F* (3, 61) = 3.91, *p* = 0.013; partial η<sup>2</sup> = 0.16) and for involvement (*F* (3, 61) = 14.16, *p* < 0.001; partial η<sup>2</sup> = 0.41). As seen in Table 2, follow-up univariate ANOVAs showed significant differences with medium-tohigh effect sizes in the frequency aspect in the home setting, and in the involvement aspect in all three settings. That is, the ADHD group was found to participate less frequently at home and was less involved in the three settings, compared to the non-ADHD group. It should be noted that no significant differences were found between the groups in the frequency aspect in the school setting.

**Table 2.** Comparison of participation patterns (frequency and involvement), desire for change, and the overall environmental support in the different settings between the two groups.


Notes: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

The combined dependent variable of the desire for change significantly differs between the two groups (*F* (3, 61) = 4.83, *p* = 0.004, partial η<sup>2</sup> = 0.19). In follow-up univariate ANOVAs, significant differences were found between the groups with medium-to-high effect sizes in the home and community settings, meaning that parents in the ADHD group were more interested in change than parents in the non-ADHD group at home and in the community, but not at school.

Examining the differences between the groups with respect to the overall environmental support, a significant difference between groups was found for the three combined settings (F (3, 60) = 13.39, *p* < 0.001, partial η<sup>2</sup> = 0.40), along with significant differences with high effect sizes that were found between the two groups in each setting. Thus, the ADHD group reported the overall environmental support to be lower than the non-ADHD group, indicating that, among the ADHD group, the environmental factors are less supportive of children's participation or are less available to lend support.

#### *3.2. Correlation between Overall Environmental Support and Participation Patterns in all Settings and for Each of the Groups*

Pearson's correlations were calculated to explore the relationships between environmental support and the participation patterns for each of the groups. As seen in Table 3, no significant correlation was found between the overall environmental support and frequency

of participation. However, significant positive correlations were found between overall environmental support and involvement of both groups in the home settings (Figure 1). In the school and community settings, significant positive correlations were found for the ADHD group, but not for the non-ADHD group (Figures 2 and 3). In conclusion, the ADHD group showed a positive and significant relationship between overall environmental support and involvement in each of the settings, whereas with the non-ADHD group a similar relationship was found only in the home setting.

**Table 3.** Pearson's correlations between environmental support and participation patterns (frequency and involvement) among children with and without attention deficit hyperactivity disorder (ADHD).


Notes: \* *p* < 0.05, \*\* *p* < 0.01; Pearson's correlations (r).

**Figure 1.** Correlations between overall environmental support and involvement among children with and without attention deficit hyperactivity disorder (ADHD) in the home environment.

**Figure 2.** Correlations between overall environmental support and involvement among children with and without ADHD in the school environment.

**Figure 3.** Correlations between overall environmental support and involvement among children with and without ADHD in the community environment.

#### *3.3. Differences in Participation Patterns and Prevalence of Environmental Factors Impacting the ADHD Group*

One-way repeated measures ANOVAs were conducted to determine whether there was a statistically significant difference in the ADHD group's participation among the different settings. Frequency of participation significantly changed with the settings as follows: *F*(2, 60) = 79.738, *p* < 0.001, partial η<sup>2</sup> = 0.73. Pairwise comparisons with Bonferroni correction showed all three settings significantly differed from one another (*p* < 0.001). The frequency of participation at home was found to be higher than in the school and the community environments, and the frequency of participation in the school was found to be higher than in the community (Figure 4).

**Figure 4.** Frequency of participation of the ADHD group in the three settings.

Involvement in participation also significantly changed with the settings as follows: *F*(2, 60) = 3.943, *p* = 0.025, partial η<sup>2</sup> = 0.12. In pairwise comparisons with Bonferroni correction, the mean of involvement at home was significantly lower than in school (*p* < 0.05), but no significant differences were found when comparing the involvement at home and school to the involvement in the community (Figure 5).

**Figure 5.** Involvement in participation of the ADHD group in the three settings.

The ADHD group's description of supports and barriers affecting participation in the different settings was evaluated by calculating the percentage of participants' consensus. Examining the supports, it appears that, in the home environment, "the attitudes and actions of therapists and other professionals" were reported as the most supportive factor (27.60%). In the school environment, the most supportive factor reported was "relationships with peers" (32.30%), followed by "the attitudes and actions of teachers, coaches, or staff" (25.80%). The result was similar for the community environment, where the most supportive factor for the ADHD group was "attitudes and actions of other members of the community" (29%).

Examining the barriers of the ADHD group, it was found that, in the home environment, "the cognitive demands" were reported as the most common inhibitors (38.70%). In the school environment, "the cognitive demands" and "the sensory stimulation" were the most frequently reported inhibitory factors (45.20% for each of them). Similar inhibitory factors were reported in the community environment, leading with sensory stimulation (19.40%) and cognitive demands (16.10%).

#### **4. Discussion**

This study was designed to examine the participation profile as affected by environmental factors (supports and barriers) among children and youth with ADHD, compared to their peers without ADHD. Additionally, the study set out to examine the relationship between overall environmental support and participation patterns (frequency and involvement) in each group, in the following three settings: home, school, and community.

#### *4.1. Comparison of Participation Patterns, Desire for Change, and Overall Environmental Supports*

Regarding the comparison of the participation patterns and the desire for change, it was found that, in the home environment, children and youth with ADHD participated less frequently and were less involved than their peers without ADHD. These findings are consistent with previous research suggesting that children with disabilities participate less frequently and are less involved at home, compared with typically developing children [16]. They also support the findings of Lavi et al. [28], showing that adolescents with ADHD participate less in the home environment than their peers without ADHD. In addition, as expected, our study found that parents of children and youth with ADHD reported a greater desire for change in their children's participation in the home environment, compared to the non-ADHD group. This suggests that the parents of children with ADHD were less satisfied with their child's participation patterns. A possible explanation for this is the difficulty in balancing the stability of the family with the need to assist a child/adolescent with ADHD [27,41], which is made even more challenging by parents' exposure to their child's ever-present difficulties in this environment.

Regarding the school setting, our findings indicated that children and youth with ADHD were less involved than children and youth without this diagnosis. This may be due to the difficulties in the social functioning of children and youth with ADHD, which comprise four of the five school-related activities in the PEM-CY questionnaire. Indeed, other studies of children with ADHD indicate high rates of peer rejection, low teacher ratings of their social skills in the classroom, difficulty in social involvement, along with difficulty in cooperation and in reciprocal conversation while playing with others [21,22]. Additionally, this finding was consistent with a previous study indicating that children and youth with disabilities (including ADHD) are less involved in the school environment than their typically developing peers [15].

However, in our study, no significant difference was found between the groups in frequency of participation in the school environment. A possible explanation for this is that participation frequency is influenced by school policy and routine [15], which may possibly obscure the differences between the groups. In addition, no significant differences were found between the groups in the desire for change in school environment participation. This may be related to the fact that the school environment is less accessible or familiar to parents.

Regarding the community environment, our findings indicated that the ADHD group was less involved than the non-ADHD group. This is similar to the study by Engel-Yeger and Ziv-On [20], which found that children with ADHD preferred to participate less in most leisure activities, and also received less enjoyment from formal leisure activities, compared to children without ADHD. Since involvement typically includes elements such as motivation and personal preference, which can be considered participation-related constructs [10], it may be assumed that the lower preference among children with ADHD to participate in the leisure activities that most often occur in the community environment would affect their involvement here. However, our study found no significant difference in frequency of participation between groups in the community environment. This finding is not consistent with the study of Bedell et al. [14], according to which children and youth with disabilities participate less frequently and are less involved in community activities than typically developing peers. A possible explanation is the difference in group characteristics, that is, the study described above included diagnoses in addition to ADHD, such as orthopedic defects, developmental delay, and autism, which may affect both the frequency and the ability to participate in activities. In addition, our study showed that parents of children and youth with ADHD reported a greater desire for change in their children's participation in the community, compared to those without ADHD. A possible explanation is that parents perceive themselves as influencing their children's participation in community activities, given their role in registering their children for such activities and encouraging them to participate.

Overall, our findings indicated lower involvement in the ADHD group compared to the non-ADHD group in all three settings. Since the construct of involvement refers to the level of concentration, emotional involvement, satisfaction, and attention when performing an activity [8], it is not unlikely that children and youth with ADHD will experience difficulty in maintaining attention and active partnership throughout a particular activity. In accordance with the lower involvement of children and youth with ADHD in all settings, the implementation of training programs for parents and teachers, as well as outreach programs in the community setting, can be beneficial for promoting their participation, personal involvement, and well-being. Regarding overall environmental support, this study found a significant difference between the groups in each of the three settings, meaning that environmental factors were less supportive of, or were less available to lend support to, the participation of children and youth with ADHD, compared to the non-ADHD group. These findings support the literature, which documents differences in the environmental support given to children with and without disabilities in different settings, and highlights the fact that there are more environmental barriers that affect the participation of children with disabilities, compared to children without disabilities [12,13]. A specific example of this is reflected in the study by Coster et al. [15], who found that parents of students with disabilities were significantly more likely to report patterns of the school environment that hindered participation, and that the resources needed to support their child's participation were not adequate, compared to students without disabilities.

#### *4.2. Relationship between Overall Environmental Support and Participation Patterns*

The results of the study found no significant association between environmental supports and participation frequency, for either of the groups (with and without ADHD). A similar finding was reported in the study of Rosenberg et al. [30], which examined the effect of environmental barriers on participation among children with mild developmental disabilities and found no significant correlation between environmental barriers and the number of activities in which the child participated (diversity) or the child's participation frequency (intensity). These results may be explained by the fact that the ADHD group can easily attend an activity, since they do not necessarily need major accessibility. However, our study did find a significant association between environmental support and involvement in all three settings among children and youth with ADHD, and in the home setting among children and youth without ADHD. It seems that, for the ADHD group, being involved (which means to be fully immersed in the activity) can be more challenging. This reinforces the understanding that participation is a multidimensional concept whose assessment requires addressing various aspects [7,9,31], including frequency and subjective measures such as involvement.

Our findings demonstrated a positive connection between environmental support and involvement among children and youth with ADHD in all the settings, confirming the need for environmental support to promote participation. In light of the importance of participation and its contribution to development, health, and well-being of children and youth, further examination is advisable to better understand the effects of increased participation on the well-being of this population.

It should be noted that, in the ADHD group, a significant correlation was found in all three settings, whereas in the non-ADHD group the correlation was found to be significant only in the home environment. These differences between groups indicate that, for children and youth with ADHD, the environmental supports are more significant influences on their involvement in activities in different settings.

#### *4.3. Comparison of Frequency and Involvement between the Different Settings and a Description of the Environmental Factors*

In comparing the different settings for frequency and involvement patterns, the home was found to be the environment in which children and youth with ADHD participated most frequently, yet they were less involved. A possible explanation for this is that the home environment is the main place where most daily tasks are performed [42]. As such, it contains activities that may take place in the family routine at a high frequency, such as household chores (e.g., setting the table and cleaning the room) and personal care management (e.g., maintaining hygiene and brushing teeth), as well as other unstructured informal activities, such as play, arts and crafts, and getting together with other people, that often require a child's initiative. At the same time, the child/adolescent may be dependent on another person, particularly his or her parents, when performing activities in this environment [34], which may result in lower involvement. An example of this was given in the study of Dunn et al. [43], who examined participation among children and youth with ADHD in various household tasks, and highlighted their need for support from family members while performing them; this tendency may also affect other activities reflected in the present study.

In contrast, school was found to be an environment where children and youth with ADHD are mostly involved in school activities, which may be related to the school activities themselves that are more structured. Our results, however, were limited to activities included in the PEM-CY questionnaire, which do not necessarily highlight the difficulties of this population in executive functions [44,45] and academic functioning [24,26]. These difficulties are underrepresented in the questionnaire items related to school activities, but are more prominent in home activities such as homework preparation, school preparation, and household chores.

According to our results, the community showed the lowest frequency of participation among children and youth with ADHD. This may result from the nature of community activities, such as group events or traveling, which may occur less frequently than home and school activities. Significantly, this low frequency was also reported by the non-ADHD group, an outcome supported by various studies that used the PEM-CY and similarly indicated highest participation frequency at home and lowest in the community [12,13,34]. However, it is important to note that these studies did not examine the frequency differences among the three settings, but rather compared groups with and without disabilities in different participation patterns.

Examining the environmental supports and barriers influencing children and youth with ADHD indicated that the factors which stood out most frequently as inhibitors, in all three settings, were the activity demands, and in particular the cognitive demands (e.g., concentration, attention, and problem-solving). This finding is consistent with the cognitive difficulties of the ADHD population related to attention, concentration [17], and executive functions, all of which impair their participation in various occupations throughout the day [28]. In addition to these, social demands (at home and school) and physical demands (in the community) were also reported as inhibiting participation among this population. The findings relate to activity demands, given the confirmed interactions between the person, the environment, and the activity [14]; therefore, changing the activity demands in the environment may promote participation. Interventions that include this adaptation of activity demands may promote the participation of children and youth with ADHD in the various settings. Adaptations in the activity demands, and in particular in the cognitive demands, can be applied by professional training and guidelines to parents, teachers, and community members who are involved in the participation of this population.

Another major barrier for children and youth with ADHD in school and the community is the sensory stimulation (e.g., amount and/or type of sound, noise, light, temperature, textures of objects, and crowds). This refers to distractions due to unrelated stimuli, and

it may therefore reflect the high prevalence of comorbidity associated with sensory modulation dysfunction [46]. Reports of sensory interference highlight the need to assess environmental sensory conditions (for example, by using the Sensory Processing Measure and/or the Sensory Profile Questionnaire) and to consider them when designing interventions, in order to promote participation. Reducing sensory stimuli such as classroom decorations, and/or performing community activities in a relatively quiet environment, may be examples of such sensory adjustments.

Regarding the supports, other people's attitudes and actions were found to promote the participation of children and youth with ADHD in the three settings (e.g., babysitters and other professionals at home; teachers and staff at school; and members of the community, such as shopkeepers and instructors). Indeed, according to the literature, positive attitudes in the community and culture can promote participation [32], with the strongest evidence for social protective factors being found in social acceptance having a positive effect on the symptoms of the disorder [38]. In addition, relationships with peers emerged as one of the strongest supports for children and youth with ADHD. The fact that this factor was found to be helpful in school was interesting, given that it is adult support through their presence and supervision that is sometimes perceived as providing confidence among children with ADHD in the school environment [47].

Given that the environment is a potentially modifiable factor, there is considerable value in identifying which features of the home, school, and community environments are barriers to participation, so that interventions can be directed appropriately [7]. Therefore, these findings that indicate specific environmental factors constituting supports (people's attitudes and actions, and relationships with peers) or barriers (activity demands and sensory stimulation) to participation of children and youth with ADHD may greatly contribute to the well-being of that population.

#### *4.4. Research Limitations and Recommendations for Further Research*

The present study has a number of limitations, for which further research is recommended. First, the study focused on a convenience sample of 65 children and youth with and without ADHD aged 6–14 years, which is a wide age range. Moreover, most of them have average or above-average socioeconomic status, live in the central district of Israel, and have parents with an academic education. Further studies need to include a larger and more representative sample of the two groups, including smaller age ranges, various socioeconomic strata and areas of residence, with varying levels of parents' education, in order to enable better generalization of the findings. In addition, further studies should include children and youth with various health conditions, following the literature related to the range of health-related characteristics among representative study samples. Second, the information was based on parents' reports regarding their children's diagnosis of ADHD and the study criteria. Further research could include additional information from a professional regarding the ADHD type and comorbidity, such as sensory modulation dysfunction, learning disabilities, and/or DCD, which is common in this population [27,46,48]. Additionally, it is recommended to include other perspectives, such as the child/youth themselves and/or others (teachers and caregivers), especially regarding settings outside the home. It should also be noted that parents with ADHD are more likely to have children with ADHD [27]; this may affect parents while answering a long questionnaire like the PEM-CY, as well as their responsiveness to participating in the study. Therefore, it is recommended to substitute or add tools, including semi-structured interviews with parents, in order to facilitate questionnaire completion and deepen their understanding of the participation patterns. Moreover, this study included one measure for evaluating participation, which is a multidimensional and complex concept that no single dimension of measurement is likely to fully capture [7]. In following up the study results, further research should use additional measures related to the more subjective patterns inherent in this concept. Finally, further examination of the environmental supports and barriers, as well as the impact of environmental supports on the involvement and well-being of

children and youth with ADHD, may also contribute to the expansion of professional knowledge.

#### **5. Conclusions**

This study described the participation profile, environmental factors (supports and barriers), and the relationship between overall environmental support and the participation patterns of frequency and involvement, among children and youth with and without ADHD, aged 6–14, based on parental reports. The findings indicated that, compared to their peers without ADHD, children and youth with ADHD participate less frequently in the home setting, they are less involved in all three settings, and their parents are more interested in changing their participation at home and in the community. Information about the specific activities to which parents want to see change is clinically important, as it can guide goal setting and targeted intervention. At the same time, children and youth with ADHD reported lower overall environmental support. Our findings showed a relationship between environmental support and involvement of children and youth with ADHD in all three settings, in contrast to children and youth without ADHD. These differences between groups reflect the interactional nature of participation, while confirming the need for environmental support to promote participation mainly among children and youth with ADHD.

In addition, differences in their participation patterns were found in various settings. These findings highlight the need for a broad examination of participation in different settings. As mentioned above, since participation is a multidimensional concept, its assessment requires addressing multiple aspects, including subjective measures such as involvement.

Different environmental factors were found to support or inhibit the participation of children and youth with ADHD, such as other people's attitudes and actions, relationships with peers, activity requirements (particularly cognitive), and sensory stimuli. This knowledge can lead to a greater effort to evaluate environmental support for children and youth with ADHD and improve their participation patterns (particularly their involvement) in various activities in different settings. The resulting development of intervention programs will benefit this population and contribute to their well-being.

**Author Contributions:** Investigation and Writing, T.S.; Statistical consulting, Writing—Data analysis and Results, H.F.-G.; Professional advisor, D.A.; Primary Investigation, Writing, and Corresponding author, A.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the ethics guidelines and approved by the Institutional Review Board (Ethics Committee of the Hebrew University, Jerusalem, Israel; No. 27032018; 2018).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The dataset generated and analyzed during the current study is available to the first and last authors, but are not publicly available due to ethical guidelines.

**Acknowledgments:** We are grateful to the parents and children who participated in this study.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


International Journal of *Environmental Research and Public Health*

## *Article* **Exploring the Participation Patterns and Impact of Environment in Preschool Children with ASD**

**Ghaidaa Khalifa 1,\*, Peter Rosenbaum 2,3, Kathy Georgiades 4, Eric Duku <sup>4</sup> and Briano Di Rezze 1,3**


Received: 1 June 2020; Accepted: 3 August 2020; Published: 6 August 2020

**Abstract:** Participation in everyday activities at home and in the community is essential for children's development and well-being. Limited information exists about participation patterns of preschool children with autism spectrum disorder (ASD). This study examines these participation patterns in both the home and community, and the extent to which environmental factors and social communication abilities are associated with participation. Fifty-four parents of preschool-aged children with ASD completed the Participation and Environment Measure for Young Children and the Autism Classification System of Functioning: Social Communication. The children had a mean age of 48.9 (8.4) months. Patterns of participation were studied using descriptive statistics, radar graphs, and Spearman correlations. Children with ASD participated in a variety of activities at home and in the community, but showed a higher participation frequency at home. Parents identified different barriers (e.g., social demands) and supports (e.g., attitudes) in both settings. There was a moderate positive association between children's social communication abilities and their levels of involvement during participation and the diversity of activities. This study highlights the importance of social communication abilities in the participation of preschool children with ASD, and the need to support parents while they work to improve their child's participation, especially within their communities.

**Keywords:** autism spectrum disorder; participation; environment; social communication; childhood

#### **1. Introduction**

Participation is defined in the WHO's International Classification of Functioning, Disability, and Health (ICF) as "involvement in a life situation" [1]. Since the introduction of the ICF, this definition has evolved and has been given several meanings in the literature [2,3]. Participation has also been described as the intensity of engagement or being involved in a life situation [2], and as the experience of taking part in an everyday activity [4]. Participation has been defined as a multidimensional concept that includes two essential constructs: *Attendance* to an activity, and level of *involvement* [5,6]. *Attendance* is defined as "being there" and is measured by the frequency and/or diversity of activities in which the person takes part [5]. *Involvement* is defined as "the experience of participation while attending, including elements of motivation, persistence, social connection, and affect" [5]. The definition by Imms and colleagues [5] informed this study, as this multidimensional concept could be applied to any activity or setting, regardless of the ability of the individual [6].

It is believed that participation is a pre-requisite for human development [7] and an indicator of children's health and well-being [8,9]. According to Bandura's social learning theory [10], new skills are acquired by direct experience and engagement with and/or through the observation of others. Therefore, through participation in everyday activities, children develop cognitive, sensory, motor, and social skills [11], form friendships, and develop their sense of self-identity [7]. Overall, participation is associated with positive outcomes for all children, but it could have more significant impact on the development of children with disabilities. Participation has been reported to have an influence on learning, independence, and social inclusion of children with disabilities [9].

Over the last decade, the number of children diagnosed with autism spectrum disorder (ASD) has increased, with 1 in 54 children diagnosed with ASD in the US [12] and 1 in 66 in Canada [13]. Children receive an ASD diagnosis during the preschool years (median age of diagnosis is 4 years) [14]. Parents are usually stressed and overwhelmed following receiving an ASD diagnosis [15] and their children's participation might not be their priority. The preschool years are the period where children first start to learn their roles in a group, gain new skills, and practice these skills in their environments [16]. In addition, participation in the preschool years highly depends on the opportunities offered to children by adults in their everyday environment, typically their parents or caregivers [17,18]. The literature indicates that children with disabilities participate less frequently in domestic, educational, leisure, and social activities when compared to their typically developing peers [11]. Children with ASD are reported to have limited participation, as well as engaging less frequently and in fewer activities when compared to their typically developing peers [19,20]. They participate less frequently in activities of self-care, community mobility, and leisure activities [19]. Families of preschool children with ASD are reported to participate less in special event activities such as family vacations and birthday parties [20]. School-aged children with ASD are also reported to participate less than their typically developed peers in social activities, unstructured activities, and after school activities [21,22].

Many challenges associated with ASD, such as social communication deficits and/or repetitive behaviors, put children with ASD at risk of limited participation. Their social communication difficulties make it a challenge to be involved and engaged with others, which is required for many aspects of participation [19,23]. Furthermore, their restrictive and repetitive behaviors may set them apart from other children and further limit their participation in everyday activities [19]. In addition, parents of children with ASD indicate that their child's participation may also be impacted because parents may avoid participation outside their home due to fears of the negative perceptions of others [24,25].

As indicated above and in the literature, various aspects of the environment-the physical, social, or attitudinal—can have a significant impact on children's participation [1,4,26]. Bronfenbrenner's bioecological systems theory identified the different layers of the environment and their impact on child development [27]. Child development is affected by their interaction with the environment at various levels (directly and indirectly), including their immediate family, community, and society [27]. This emphasizes the need to look at the potential impact of various environments to understand children's development. Parents of children with disabilities consider the environment to be less supportive and believe that their children have more environmental barriers than typically developing children [9,28]. These barriers could relate to the physical, social, or attitudinal aspects of the environment. The environmental features may either support or hinder the participation of children with ASD. Some of these features include sensory issues, such as level of noise and lighting [29]; furthermore, the physical layout of the space, as well as the social and cognitive demands of some activities, may compromise social connections, such as interacting with others [29]. Availability of resources and services may also support participation of children with ASD [29].

Studies of the impact of the environment on participation with various populations of children with disabilities have shown inconsistent findings. In their study of participation of school-aged children with severe physical disabilities, King et al. [30] found that the environment indirectly impacted on participation. For example, unsupportive environments (e.g., inaccessible or less accommodating) were found to be related to a child's reduced functional ability and therefore were associated with limited participation [30]. A study of participation of children with cerebral palsy found that environmental factors failed to predict the child's participation diversity [31]. Moreover, for preschool children with mild developmental disabilities, environmental factors were found to be significant predictors of children's participation [11]. Studies of participation of school-aged children with ASD found the environment to be one of the factors that impacted their participation [26,28]. In their report, Askari and colleagues [26] reviewed the literature on the impact of the different aspect of the environment (i.e., physical, social, and attitudinal) on participation of children with ASD. In this work, social supports from parents, siblings, or friends were highlighted as important for participation, whereas negative attitudes in the community (e.g., church) presented as barriers for participation for children with ASD [26].

Few measures are designed to assess characteristics of the environment that impact participation in different settings. These include the Child and Family Follow-up Survey (CFFS) [32] and the Participation and Environment Measure (PEM) [33]. The CFFS is designed for children 5 years and older with traumatic brain injuries (TBIs) [32]. It has five sections, one of which is the Child and Adolescent Scale of Participation (CASP), to report on the participation of children with TBIs in the home, school, and community [32]. Another scale is the Child and Adolescent Scale of Environment (CASE), which measures the intensity of the physical, social, and attitudinal environment problems experienced by children with TBIs [32]. The PEM has two versions: The Young Children Participation and Environment Measure (YC-PEM) [34] for children aged 0–5 years, and the Participation and Environment Measure for Children and Youth (PEM-CY) for children and youth aged 5–17 years old [33]. They are used to report on participation and the quality of the environment in various activities in three contexts: At home, daycare/school, and community. PEM can be used with children with various disabilities, as well as children without disabilities.

Another factor that affects participation is the child's social functioning [35,36]. Social communication functioning is inconsistently defined in the literature [37]. New perspectives in the field are making the distinction between social deficits, impairment, functioning, and abilities [38]. In an extensive search of the literature, King and colleagues developed a conceptual model of factors affecting participation in recreational and leisure activities for children with disabilities [39]; the child's social functioning was one of the factors identified in this model. Evidence indicated that better-developed social functional ability was associated with better involvement when participating in activities [39]. The Diagnostic and Statistical Manual of Mental Disorders (DSM-5) defines social communication impairment as "deficits in social-emotional reciprocity, non-verbal behavior, and imitative and make-believe play" [40]. Although deficits in social functioning are one of the core symptoms of ASD [39,41], to our knowledge there is a paucity of research on the impact of social functioning on participation for children with ASD.

To date, studies on participation of children with disabilities have focused on school-aged children and adolescents and those with physical disabilities, while there is a lack of research on participation for young children with ASD [24]. A systematic review by Adair et al. [6] found that of the 394 articles on participation that they reviewed, 105 articles focused solely on cerebral palsy, while only 37 articles focused on ASD. Furthermore, these types of studies usually involve comparing a group of children with disabilities to a group of children without disabilities [29,42–45]. There is a need to study, in depth, the patterns of participation and the potential factors associated with participation amongst preschool children with ASD. The aims of this study were to explore the patterns of participation for preschool children with ASD (3–6 years old) and investigate the impact of different environmental and individual factors on their participation.

#### **2. Methods**

#### *2.1. Participants and Procedures*

This cross-sectional study investigated the patterns of participation in preschool children with ASD and the factors that are associated with them, including the environment and the social communication abilities of the child. The study involved analysis of data relating to a subsample of children who were recruited for a larger project (the Pediatric Autism Research Cohort (PARC) project-pilot phase). The subsample included children who have completed the YC-PEM, and therefore involved children who were 5 years and younger. PARC is a longitudinal inception cohort of children recently diagnosed with ASD from Hamilton, Ontario. The study was approved by the local research ethics board (Hamilton Integrated Research Ethics Board (HiREB)) and all families provided informed consent. The sample included 94 children diagnosed with ASD. The inclusion criteria for participants involved being under age 6 at enrollment and being enrolled in services at the regional autism program.

#### *2.2. Assessment Measures*

Sociodemographic questionnaire: This questionnaire was created specifically for the PARC study and included questions about the child, such as their age, sex, and country of birth. It also asked questions concerning the family background, including their educational level and family income.

#### *2.3. Participation and the Environment*

Participation and Environment Measure (Young children version—YC-PEM): YC-PEM was developed based on ICF concepts and is a parent- or caregiver-completed questionnaire. The current study explored participation at home and community settings only. Home and community are considered the natural learning environment of daily activities for young children [46]. For each activity, parents reported: (i) Frequency of participation on an 8-point scale from never (0) to daily (7) for 13 age-appropriate activity items at home and 11 items in the community; (ii) level of involvement in specified activities (5-point scale from minimally involved (1) to very involved (5)); and (iii) whether caregiver/parent would like to see changes in their child's participation in this type of activity (yes or no question). For each setting, parents reported on various features of the environment or resources and their impact on their children's participation, such as the sensory qualities of the environment or the cognitive demands of an activity. For each item in the environment, parents chose one of the following: Whether it has no impact, usually helps, sometimes helps, sometimes makes harder, or usually makes harder. Since the study aim was to explore the pattern of participation, questions on caregivers' desire to change participation were not considered.

The YC-PEM has shown sound psychometric properties with children with different disabilities. It has an acceptable internal consistency (>0.70) for three scales: Frequency (α = 0.72); Involvement (α = 0.80); and Environmental Support (α = 0.92) [47]. The test–retest reliability for the frequency scale was fair to good for home (ICC = 0.61–0.63) and community (ICC = 0.55–0.63), and for the level of involvement scale reliability was good to excellent for the home (ICC = 0.79–0.93) and good for the community (ICC = 0.71–0.97). The reliability for the environment scale was good for the home and community (ICC = 0.91–0.94) [47]. PEM-CY/YC-PEM has been used to investigate the pattern of participation for children with ASD in different settings, but mostly for school-aged children [29,42,48].

#### *2.4. Social Communication Functioning*

The construct of social communication was explored using the Autism Classification System of Functioning: Social Communication (ACSF:SC) [49]. The ACSF:SC is a strength-based tool that aims to categorize children with ASD who are between 3 to 6 years old into one of five levels of functioning based on their social communication abilities. This descriptive tool was developed by CanChild researchers based on ICF concepts. The social communication abilities range from level V (lowest ability) through level I (highest ability). This classification tool is not meant to replace any diagnostic or assessment tools, but rather provides a simple standardized method to classify the child's social communication abilities in a consistent manner among the health provider teams, teachers, and parents [50]. A rater who is familiar with the child is asked to provide two ratings: The child's capacity level (what the child can do at their best) and the child's typical performance level (what the child can do on a day-to-day basis). The ACSF:SC demonstrates good intra-rater agreement for parents (kw = 0.61–0.69) and good to very good for professionals (kw = 0.71–0.95) [50]. The inter-rater agreement among parents and professionals ranges from fair to moderate agreement (kw = 0.33–0.53) [50].

#### *2.5. Data Analysis*

The data for the current analysis were drawn from the initial time point from the larger PARC study. Data were analyzed using STATA software, version 13 (StataCorp LLC, Texas, TX, USA), and an effect was considered statistically significant at α = 0.05.

Descriptive statistics, including the means, standard deviations, and percentages of child characteristics and their family's sociodemographic information were first calculated for the participants. The distribution of the sample among the five levels of the ACSF:SC was obtained for the best capacity and typical performance scales. To understand the pattern of participation for our sample, the mean and standard deviation were calculated for the frequency and level of involvement scales of the YC-PEM. The percentages of activities in which the children participated were also calculated. Radar graphs were obtained to illustrate the distribution of scores across items. Radar graphs are used to represent the data visually in order to examine patterns of activity and are shaped like histograms. The radar graphs have multiple spokes spreading from the center of the graph, and the longer the spoke, the higher the magnitude of the variable represented by this spoke [51].

To explore the relationships between the ACSF:SC levels and YC-PEM-reported frequency, level of involvement, and the percentage of activity for both settings, scatter plots were created to visualize the data, followed by Spearman's correlation analysis. The same procedure was done to explore the relationships among the ACSF:SC levels and the environmental scales of the YC-PEM, followed by analysis of variance (ANOVA). ANOVA was conducted to explore whether the size of the differences between best capacity and typical performance levels was associated with the presence of environmental supports or barriers.

#### **3. Results**

Descriptive statistics: 54 children completed the ACSF:SC and were included in the analysis. Socio-demographic information of the parents, their household, and their child with ASD is summarized in Table 1.


**Table 1.** Descriptive statistics and sociodemographic features of the sample.

Non-respondent analysis: Six participants did not complete the ACSF:SC and were excluded from the study. There were no significant differences between the respondent and non-respondent children in terms of their age (t (58) = 0.15, *p* = 0.9), gender (Pearson X<sup>2</sup> (1) = 1.0, *p* = 0.3), or language spoken at home (Pearson X<sup>2</sup> (1) = 0.36 *p* = 0.6).

ACSF:SC best capacity and typical performance scores: Parents of 50% of the participants rated their child the same for typical performance and best capacity, and 44.4% of parents judged their children to have lower typical performance abilities than their best capacity ability. Parents of only 5.6% of the participants judged their children to have higher typical performance abilities than their best capacity (Table 2). A total of 46.3% of participants had a ±1-level difference, while only 2% had a 2-level difference.


**Table 2.** Autism Classification System of Functioning: Social Communication (ACSF:SC) best capacity and typical performance ratings. Agreement between best capacity and typical performance is highlighted.

#### *YC-PEM*

Participation: Overall, parents reported their children as participating in a variety of activities at home and in the community. Frequency and level of involvement are demonstrated in the radar graphs to depict the activities in which children engaged within the home and community settings (Figures 1 and 2).

**Figure 1.** *Cont*.

**Figure 1.** (**a**) Mean frequency of activity participation at home, (**b**) mean frequency of activity participation in the community.

**Figure 2.** (**a**) Mean level of involvement in activities at home, (**b**) mean level of involvement in activities in the Community.

#### **4. Home Setting**

Activity frequency and level of involvement: The majority of our sample (>73%) were reported to participate most frequently in basic care routine activities (mean = 6.6) (Figure 1a); 50% were reported to be "somewhat involved" in these activities (Figure 2a). Household chores were reported to have the lowest frequency in the home setting with a mean of 1.8 ("few times in the last four months") in three out of four activities. Participants were reported to have different levels of involvement ranging from "not very involved" to "very involved". Furthermore, up to 65% of the participants were reported to have never participated in these chores. Participants showed high frequency rates in the interactive and organized play with the majority (98%) participating daily in these activities and being very involved. Socializing with friends and family was also reported to have low frequency from "few times in the last four months" (44%) to "a few times a month" (up to 27%). The level of involvement ranged from not very involved to very involved (Supplementary Table S1).

Environmental supports and barriers: Half of the parents reported that the physical layout of their houses supported their children's participation, as shown in Table 3. Sensory qualities were perceived as a support for 37.7% of the parents. Cognitive and social demands were reported to support children's participation for 22.6% and 26.4%, respectively, with a further 24.5% and 30.2% parents considering them as barriers. The attitudes of family were reported as supportive for 34.7% of parents. A total of 46.7% of parents considered money and time to support their children's participation.

Relationship between social communication and participation: Spearman's correlation analyses provided the same correlations for best capacity and typical performance levels and participation. Therefore, we decided to use the typical performance levels as they represent everyday functional performance. There was very low correlation between participation *frequency* and the ACSF:SC (typical performance level) (r = −0.02, *p* = 0.9). However, the Spearman's rank correlation showed a low negative correlation between the *level of involvement* and the ACSF:SC (r = −0.32, *p* < 0.01), and a moderate negative correlation between the *percentage of activity participation* and the ACSF:SC (r = −0.42, *p* < 0.01). Because of the scaling of the ACSF:SC, the correlation was negative, but there was a positive association between the level of involvement, percentage of activity participation, and social communication (i.e., the better the social communication ability on the ACSF:SC, the higher the level of involvement and the wider the variety of activities in which the child participates).


**Table 3.** Environmental features as perceived by parents at home.

#### **5. Community Setting**

Activity frequency and level of involvement: Overall, the frequency of participation in the community activities was lower than those in the home setting (Figure 1b). Children were reported to participate most frequently in shopping and errands (once a week), followed by unstructured physical activities (a few times a month). The lowest frequency observed was 1 (once in the last four months) for classes and lessons, organized physical activities and overnight trips, vacations and visits. However, even with the low frequency, children were reported as being involved when doing these activities. In all of the activities, there were parents who reported that their children never participated in these activities. For example, 73.4% of parents reported that their children never participated in an organized physical activity.

*Environmental supports and barriers*: The sensory quality of the environment was reported by parents as a barrier for 23.1% of the participants (Table 4). Cognitive and social demands were reported

by parents as supportive for 22% and 11.8% of the participants, respectively, while reported as barriers for 24% and 35.3% of the participants, respectively. Parents reported attitudes and relationships with friends to be supportive for 25% and 24% of the participants, respectively. Personal transportation, equipment, and supplies were reported as supports for more than 60% of participants. Time and money to support their children's participation at the community were also reported by 40% of the parents.

Relationship between social communication level and participation: There was very low correlation between the ACSF:SC (typical performance) level and the participation *frequency* or the percentage of activity participation. There was a moderate negative correlation between the ACSF:SC (typical performance) level and the *level of involvement* (r = −0.41, *p* = < 0.01). Once again, although the correlation was negative, there was a positive association between the level of involvement and social communication abilities (i.e., the better the social communication ability on the ACSF:SC, the higher the level of involvement when children participate in activities).

ACSF:SC and the environmental supports and barriers: The Spearman's analysis showed very low correlation (r ≤ 0.03) between the ACSF:SC levels and the environmental supports or barriers for both the home and community settings.

The sample was then divided into three groups based on the size of differences between the best capacity and typical performance levels of the ACSF:SC. In group 1, differences were ≤−1, in group 2 there were no differences, and in group 3 the differences were ≥+1. The ANOVA showed no difference between the groups in terms of home environmental support (F (2, 50) = 0.08, *p* = 0.9). Since the environmental barriers at home, as well as environmental supports and barriers in the community were not normally distributed, the Kruskal–Wallis tank test was conducted but no difference was found between groups (*p* > 0.05).


**Table 4.** Environmental features as perceived by parents in the community.

#### **6. Discussion**

This descriptive study explored participation patterns of preschool children with ASD and factors associated with participation, including the environment and the social communication abilities of the child.

Participation pattern for preschoolers with ASD in different settings: Overall, preschool children with ASD participated in a variety of activities at home. Organized play activities, such as screen time, indoor play, and games, were reported to have the highest frequency, which was also reported in other studies [19,29,42,52]. In fact, in one study, children with ASD had a higher frequency of participation than their typically developed peers in activities such as watching TV and screen time [29]. These activities usually do not involve socializing or engaging with others. Previous studies found

that children with ASD usually participate in activities alone or with few people—usually their families [22,42].

Children in this study were also reported to have the lowest frequency of participation in household chores. For example, the mean frequency of participation in meal preparation was 1.8 (out of 7), for taking care of family members was 1.8, and for laundry and dishes was 1.7. These findings were also evident in the literature with preschool and school-aged children with ASD [19,53]. When asked, parents revealed that they did not consider assigning chores to their children with ASD [19]. Parents reported that offering chores to their children with ASD would require a lot of energy to accommodate their children's behaviors and needs, and therefore they chose not to engage them in these activities [19]. Participating in chores could provide children with ASD with the opportunity to practice their problem-solving skills, increase family socializing, teach them to take responsibilities, and prepare them to take care of themselves and others [54,55].

Our findings also indicated that children with ASD generally have lower rates of participation in community settings (mean = 2.9) when compared to a home setting (mean = 5.9). The same findings were reported for children with various disabilities [56,57]. Parents of children with ASD reported having less control over the environment in the community [11,53]. It is more challenging for parents to manage their children's behavior in the community due to the unpredictability of the situations and sensory stimulation. As such, families reported that their energy is spent trying to think about the environment—what to expect and how their child may react [19]. The whole process is exhausting for them and consequently they avoid participating in activities in the community [53]. When considering participation for children with ASD in the community, this highlights the importance of taking the whole family into consideration as a unit, rather than only focusing on the child and their capabilities. These findings are consistent with Bronfenbrenner's Ecological Theory of Human Development [27], in which the child's developmental outcome is influenced by their interactions with different levels of the environment. At the level of the microsystem, child development is influenced by their immediate environment, which typically includes the family [27]. Parents are responsible for offering opportunities for their children to participate [19,53]. In one study, parents reported avoiding dining out or taking their child to grocery stores because of their risk of a behavioral meltdown [53]. This is supported by our findings that even though these children have generally lower frequency of participation in the community, some were reported to have a high level of involvement of participation in activities in the community. For example, participating in overnight trips and vacations had the lowest frequency (1 out of 7) in the community; however, children who participated had the highest level of involvement (3.8 out of 5) compared to all other community activities. Although there were no control groups in the current study to see how patterns of participation in this cohort compare to those of children without ASD, other studies have reported common findings that children with disabilities have lower participation frequency and involvement than children without disabilities [26,28]. Even when children with and without disabilities participated in the same activities, their levels of involvement are different [26].

Environmental barriers and supports: Parents reported a variety of environmental supports and barriers. However, in some cases what was reported as a support for some parents was considered as a barrier for others. For example, 22% of parents considered the cognitive demands of an activity as a support to their child's participation at home; however, the same percentage of parents considered it as a barrier. The same applies for social demands of the activity and the availability of services, where similar percentages of parents had considered it as either a support or a barrier. This underscores the importance of taking into consideration the individual variations among children with ASD and how the needs of each child vary in different contexts [28]. Furthermore, sensory qualities of the environment were considered mainly as a support at the home, while a higher percentage considered it as a barrier in the community. This lends further support to the fact that parents' lack control over the community environment and its impact on their children's participation. It also supports the findings of another study where atypical sensory processing, such as hyper-responsiveness, was associated with lower frequency of activity participation in the community [58].

Relationship between social communication and participation: Our findings indicated that better social communication abilities were associated with a wider variety of activities in which the child participated at home, and higher levels of involvement when participating in these activities. However, in the community, better social communication abilities were only associated with higher level of involvement, which aligns with the findings that identify the complexity of participation in the community and the different factors that impact it.

#### *Clinical Implications*

The study findings provide some insights for clinicians who work with children with ASD and their families. One important implication for service providers is actively to encourage parents of preschoolers to involve their children in as wide a range of daily activities and recreational opportunities as possible from a very young age, so that "participation" becomes part of daily life and is not then seen as a prescribed "add-on". Young children's participation in activities is a reflection of their family choices, available opportunities, as well as their abilities and interest. Whereas typically-developing children often take the initiative to be involved in activities, families of children with ASD need to be supported and encouraged to see opportunities to help improve their children's participation at home and in the community. Considering each child's individual needs, clinicians could provide some strategies to improve their participation. For example, household chores could be modified and broken down into several steps that the child could follow to improve various skills, such as their problem-solving skills. Clinicians could also provide some strategies to manage children's behavior in the community to increase their participation. When recommending interventions, clinicians need to take into consideration the family as a whole and any special situations they might have. Clinicians should also be aware of the community with regard to sensitivities, needs, and vulnerabilities of children with ASD.

#### **7. Limitations and Future Research**

Results of this study should be considered in light of possible sampling and data limitations, including the small sample size and the study design. Cross sectional data limits our ability to identify whether social communication ability increases participation or whether the opposite is true. However, the PARC study continues to collect longitudinal data, which will provide the opportunity to further examine this in the future. In addition, only families who are enrolled in ASD services were included, which could be a potential source for sample selection bias (access to service bias). For example, children being seen could have complex issues while children with higher cognitive abilities may not be seen within the clinical setting. Furthermore, this study was based on parents' recall and no direct observation of children's participation was conducted. Future studies could include a qualitative dimension for a deeper understanding of children's participation from parents' perspectives. Other factors that may impact on participation, such as socioeconomic status and maternal education, could also be investigated in future studies. Participation patterns could also be investigated longitudinally in future studies. Simpson and colleagues (2019) studied longitudinally the participation pattern of children with ASD who are transitioning to adolescents (age 9 and 10 years old) [48]. Over three years, they found a trend regarding socializing and participating in physical activities (participation declined as children's ages increased) [48]. Similar studies with different age groups are essential and would highlight the important factors to consider for intervention planning to improve or maintain their participation in various activities.

#### **8. Conclusions**

This study adds to the emerging body of literature on participation patterns for preschool-aged children with ASD. In addition, it explores the relationship between social communication and participation, which is a key factor central to ASD. Preschool children with ASD participated in various activities at home and in the community, which are the main environments for participation for this age group. However, parents need support to facilitate and improve their children's participation in both settings. Furthermore, for interventions to be successful, especially those intended to modify the environment, the individuality of children with ASD, with variable abilities, should be acknowledged and considered when planning intervention goals. In addition, interventions should go beyond modifying the environment around the children and consider the environments that support them, including their family.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1660-4601/17/16/5677/s1, Table S1: Percentage of children participation frequency and level of involvement.

**Author Contributions:** G.K. and B.D.R. conceptualized the idea and methodology for this study. Data curation, investigation, and formal analysis of the data were conducted by G.K. and validated by E.D., G.K. was responsible for the original draft preparation. B.D.R., P.R., K.G., and E.D. were responsible for reviewing and editing the manuscript, and B.D.R. supervised the project. All authors have read and agreed to the published version of the manuscript.

**Funding:** Funding support for this study was provided by the following: The Hamilton Health Science Research Early Career Award (ECA) 2018–2020; the Hamilton Health Sciences New Investigator Fund (NIF) 2019.

**Acknowledgments:** The authors thank all the children and families who participated in the PARC Project. The authors also acknowledge Stelios Georgiades and Anna Kata from the PARC Project Team for their support, as well as the research staff members and trainees who contributed to this study. This study was supported by the Faculty of Health Sciences and Department of Psychiatry & Behavioural Neurosciences at McMaster University, McMaster Children's Hospital Research Collaborative, Hamilton Health Sciences, and by an award from the Autism Spectrum Disorders Research Project of Grand Master Paul E Todd-2017–19. At the time of writing, G.K. was supported by a scholarship from King Saud University for Health Sciences, Jeddah, Saudi Arabia.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Concept Paper* **Definitions and Operationalization of Mental Health Problems, Wellbeing and Participation Constructs in Children with NDD: Distinctions and Clarifications**

**Mats Granlund 1,2,\*, Christine Imms <sup>3</sup> , Gillian King 4, Anna Karin Andersson 1,2, Lilly Augustine 2,5 , Rob Brooks <sup>6</sup> , Henrik Danielsson 2,7, Jennifer Gothilander <sup>8</sup> , Magnus Ivarsson 2,7 , Lars-Olov Lundqvist 2,9 , Frida Lygnegård 1,2 and Lena Almqvist <sup>8</sup>**


**Abstract:** Children with impairments are known to experience more restricted participation than other children. It also appears that low levels of participation are related to a higher prevalence of mental health problems in children with neurodevelopmental disorders (NDD). The purpose of this conceptual paper is to describe and define the constructs mental health problems, mental health, and participation to ensure that future research investigating participation as a means to mental health in children and adolescents with NDD is founded on conceptual clarity. We first discuss the difference between two aspects of *mental health problems*, namely mental disorder and mental illness. This discussion serves to highlight three areas of conceptual difficulty and their consequences for understanding the mental health of children with NDD that we then consider in the article: (1) how to define mental health problems, (2) how to define and assess mental health problems and mental health, i.e., wellbeing as separate constructs, and (3) how to describe the relationship between participation and wellbeing. We then discuss the implications of our propositions for measurement and the use of participation interventions as a means to enhance mental health (defined as wellbeing). Conclusions: Mental disorders include both diagnoses related to impairments in the developmental period, i.e., NDD and diagnoses related to mental illness. These two types of mental disorders must be separated. Children with NDD, just like other people, may exhibit aspects of both *mental health problems* and wellbeing simultaneously. Measures of wellbeing defined as a continuum from flourishing to languishing for children with NDD need to be designed and evaluated. Wellbeing can lead to further participation and act to protect from mental health problems.

**Keywords:** concept; mental health problems; mental health; wellbeing; participation; concept

**Citation:** Granlund, M.; Imms, C.; King, G.; Andersson, A.K.; Augustine, L.; Brooks, R.; Danielsson, H.; Gothilander, J.; Ivarsson, M.; Lundqvist, L.-O.; et al. Definitions and Operationalization of Mental Health Problems, Wellbeing and Participation Constructs in Children with NDD: Distinctions and Clarifications. *Int. J. Environ. Res. Public Health* **2021**, *18*, 1656. https:// doi.org/10.3390/ijerph18041656

Academic Editor: Paul B. Tchounwou Received: 2 December 2020 Accepted: 2 February 2021 Published: 9 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

Children with impairments are known to experience more restricted participation than other children [1]. It also appears that low levels of participation are related to a higher prevalence of mental health problems in children with neurodevelopmental disorders (NDD) [2,3]. NDD is a group of early onset conditions associated primarily with the functioning of the neurological system and brain, including diagnoses such as attentiondeficit/hyperactivity disorder, autism, and intellectual disability [4]; sometimes, cerebral palsy is also seen as an example of NDD, although it is primarily presented as a motor disorder in DSM-V. NDDs lead to impairments in physical, social, or academic functioning, which affect different aspects of participation.

In the Family of Participation Related Constructs (fPRC) framework, participation is described as consisting of two dimensions: physical or virtual attendance in activities, which is seen as a necessary prerequisite for the second dimension, involvement while attending the activity [5]. The fPRC framework builds on the International Classification of Functioning, Disability, and Health (ICF) definition of participation [6] by specifying two separate dimensions of attendance and involvement. It has been suggested that participation is a determinant of mental health [7]; however there is not currently a deep understanding of the relationships between mental health and participation within NDD.

While a higher prevalence of mental health problems is reported for children with impairments [3,8], especially for children with NDD, the suppositions behind the higher prevalence are implicit rather than explicit. In medical literature, mental health is commonly defined as the absence of mental health problems [9], but without a clear definition of the construct of mental health being provided. Based on Jahoda [10], Westerhof and Keyes [11] suggest that mental health is a positive phenomenon that is more than the absence of mental health problems. They define mental health in terms of hedonistic and eudaimonic wellbeing, which is also the definition we will defend in this article.

Children and adolescents with NDD seldom receive non-pharmacological mental health interventions specifically aimed at reducing mental health problems [12], although studies aimed at increasing subjective wellbeing with the help of mindfulness intervention exist [13]. We suggest that participation interventions—that is, those that aim to improve attendance or involvement in varied life situations—can be implemented to strengthen mental health as well as indirectly prevent or decrease mental health problems in children and adolescents with NDD. Thus, interventions aimed at increasing participation may be a means to increase perceived mental health [14,15]. To test this proposal, the conceptual relations between mental health problems, mental health (defined as wellbeing), and participation need to be clarified. The purpose of this paper is to describe and define these constructs to ensure that future research investigating participation as a means to mental health is founded on conceptual clarity.

To achieve our purpose, we first discuss the difference between two aspects of mental health problems, namely mental disorder and mental illness. This discussion serves to highlight three areas of conceptual difficulty and their consequences for understanding the mental health of children with NDD that we then consider in the remainder of this article: (1) how to define mental health problems (and delimit them from mental disorders and mental illness), (2) how to define and assess mental health problems and mental health (i.e., wellbeing) as separate constructs, and (3) how to describe the relationship between participation and wellbeing. We then discuss the implications of our propositions for measurement and the use of participation interventions as a means to enhance mental health (defined as wellbeing), thus proposing a way forward.

#### **2. Issues of Classification of Mental Disorders and NDD in Diagnostic Manuals**

In the ICF, aspects of functioning disability and health are classified as body structure and function, activity, and participation, thereby building on a bio-psycho-social model. The ICF is supposed to be a supplement to the diagnostic manuals used in medicine disorders, the International Classification of Diseases (ICD-11) [4], and mental disorders,

problems

the Diagnostic Systems Manual (DSM 5) [16]. These diagnostic systems include NDD, for example, intellectual disability and ADHD, within the classification of types of mental disorder, along with schizophrenia, depression, and disorders due to substance abuse [6]. In the ICD-11, mental disorders are defined and described in chapter 6: Mental, behavioral, or neurodevelopmental disorders. This chapter states:

"*Mental, behavioral and neurodevelopmental disorders are syndromes characterized by clinically significant disturbance in an individual's cognition, emotional regulation, or behavior that reflects a dysfunction in the psychological, biological, or developmental processes that underlie mental and behavioral functioning. These disturbances are usually associated with distress or impairment in personal, family, social, educational, occupational, or other important areas of functioning.*" (Chapter 6, p.1 ICD-11, 2020)

In this description, the relationship between mental disorders and everyday functioning is emphasized. The definition provides core aspects to look for when diagnosing mental disorders (cognitive, emotional, or social abilities and behavior), but it is only in the subclassifications that a distinction is made between NDD and mental illnesses such as depression or general anxiety disorder. In discriminating between different mental disorders, the ICD-11 states that NDD is characterized by symptoms that emerge in the developmental period; however, this characteristic is not unique to NDD, as other mental disorders may also present in the developmental period.

Because classification systems like ICD-11 and DSM-V are designed to define "disease" or "condition", they do not define positive mental health and do not explicitly make a distinction between bio-psycho-social levels. Unless outcomes in terms of mental health are clearly defined, it is difficult to assess mental health other than as the absence of a disease or condition. Unless mental health outcomes are clearly defined, it is difficult to plan interventions aimed at improving mental health for children and adolescents with NDD, because no positive outcome other than lack of mental health problems is described. In Table 1, definitions of the terms used in this paper are presented.

Mental disorder Mental, behavioral, and neurodevelopmental disorders are syndromes characterized by clinically significant disturbance in an individual's cognition, emotional regulation, or behavior that reflects a dysfunction in the psychological, biological, or developmental processes that underlie mental and behavioral functioning. These disturbances are usually associated with distress or impairment in personal, family, social, educational, occupational, or other important areas of functioning. ICD 11, version 09/2020, chapter 6 (http://id.who.int/icd/entity/334423054) [4] Neurodevelopmental disorder Neurodevelopmental disorders are behavioral and cognitive disorders that arise during the developmental period that involve significant difficulties in the acquisition and execution of specific intellectual, motor, language, or social functions. Although behavioral and cognitive deficits are present in many mental and behavioral disorders that can arise during the developmental period (e.g., Schizophrenia, Bipolar disorder), only disorders whose core features are neurodevelopmental are included in this grouping. The presumptive etiology for neurodevelopmental disorders is complex, and in many individual cases is unknown. ICD 11, version 09/2020, chapter 6 (http://id.who.int/icd/entity/1516623224) [4]

**Table 1.** Definitions of key terms.


fulfilling criteria for a diagnosable mental illness [9]


#### **Table 1.** *Cont.*

#### **3. Core Difficulties with the Definition and Operationalization of the Constructs Defined for Children with NDD**

We identify three core difficulties with definitions and operationalization of these constructs:

*Problem 1*. How to define mental health problems in children with NDD and delimit them from mental disorders and mental illness.

*Problem 2.* How to define and assess mental health problems and wellbeing as separate constructs in children with NDD.

*Problem 3.* How to describe the relationship between participation and wellbeing in children with NDD.

#### *3.1. Problem 1: Distinguishing Mental Health Problems from Mental Illness and Mental Disorders in Children with NDD*

We propose that any construct and measure of mental health problems should be equally applicable for children and adolescents, regardless of cognitive and physical impairment, and without defining mental illness as equal to the impairment

The constructs mental disorders, mental illness, and mental health problems all concern problems related to mental function. In this section, we first discuss the relationship between mental disorders and mental illness and thereafter the relationship between mental illness and mental health problems.

#### 3.1.1. Mental Disorders and Mental Illness—The Example of NDD

The World Health Organization [6] provides a general definition of a health condition as: "an umbrella term for disease (acute or chronic), disorder, injury or trauma. A health condition may also include other circumstances such as pregnancy, ageing, stress, congenital anomaly, or genetic predisposition" [6], (p. 228). The definition of a mental disorder seems to build on the general definition of a health condition [4]. Mental illness is not formally defined in the ICD-11 but is, in everyday language, used to describe mental

disorders other than NDD, such as mood disorders, anxiety, and fears [9]. In the ICD-11, NDD diagnoses are included as examples of mental disorders. The definition of NDD (see Table 1) stresses that NDD concerns cognitive and behavioral problems that arise during the developmental period. This implies that NDD has qualities distinct from other mental disorders and is, therefore, a sub-category of its own. One could argue, however, that there is a good case for not separating different diagnoses based on whether they indicate cognitive difficulties or not, since, for example, cognitive impairments are also a part of the clinical picture in severe depression (in that case time-limited), schizophrenia (more permanent in nature), and addiction. The fact that NDD primarily concern intellectual, motor, and/or social functions that are more or less permanent, compared to mental health problems, and arise during the developmental period provides an argument for separating NDD from mental disorders that can be described as mental illness.

#### 3.1.2. Mental Illness and Mental Health Problems

In their conceptual analysis of mental health, mental disorders, and mental health problems, Bremberg and Dalman [9] illustrate the overlap between the constructs using a figure. We have adapted their figure by making a distinction between mental disorders and mental illness to illustrate our argument (made above) that NDD does not necessarily involve mental illness (see Figure 1).

**Figure 1.** Relations between different concepts used when discussing mental health.

As Figure 1 illustrates, in many cases mental health problems overlap with wellbeing: mental health problems are a normal part of people's lives, but so is wellbeing. However, mental health problems also partly overlap with mental illness: having persistent mental health problems in childhood increases the probability of being diagnosed with a mental illness in adulthood [21]. Figure 1 also shows that mental illness is completely subsumed within mental health problems, but some mental disorders (e.g., NDD) do not automatically overlap with mental illness or mental health problems.

#### 3.1.3. Difficulties in Defining and Operationalizing Mental Health Problems Following from Conceptual Diffuseness

When definitions of constructs or diagnoses, such as mental health and NDD, are restricted to separate and different levels of the bio-psycho-social model, there is no risk for confusion or overlap, e.g., between traumatic brain injury, which is defined primarily on basis of etiology on the biological level, and behavior problems (as measured by CBCL). However, the risk for confusion between symptom-based diagnoses, such as NDDs, and mental health constructs operating at the same level(s) of the model, is more probable.

An example of a very practical consequence of the conceptual overlap between a mental disorder and a mental health problem, which may create conceptual confusion, is how authors define mental health problems when screening children with NDD. A study by Bailey et al. [22] used two indexes of mental health difficulties, as suggested for typical populations [23], namely internalizing (emotional and peer problems) and externalizing (conduct and hyperactivity) mental health difficulties, based on the definition of a mental disorder. The same type of indices is used with other "problem behavior" screening instruments, such as the Child Behavior Checklist (CBCL) [24]. Using this operationalization, children with intellectual disability have significantly more mental health problems than typically functioning children. However, this fusion of what may be factors related to cognitive impairments (i.e., communication/peer problems and hyperactivity) and behavior problems (i.e., emotion and conduct problems) may in fact lead to an overestimation of the prevalence of externalizing and internalizing mental health problems among children and adolescents with diagnoses of NDDs.

Longitudinal studies of behavior problems involving children and adolescents with diagnosed NDD [21] and children with self-reported NDD problems [25] suggest that there is a continuum of chronicity for common mental disorders. These studies suggest that problems not necessarily related to mental illness but to consequences of cognitive impairments (hyperactivity and peer problems) have stronger stability over time than mental disorders that can be described as mental illness (i.e., anxiety and depression).

Mental illness is seen as a severe and intensive type of mental health problem, situated completely within the broader circle of mental health problems (see Figure 1). The point at which mental health problems are severe enough to be diagnosed as a mental illness is debatable and somewhat arbitrary [9]. Most mental health problems have a shorter duration and less severity than a mental illness. Mental illnesses are primarily identified through diagnostic interviews where the person is required to meet certain criteria regarding the severity and persistence of problems to receive a diagnosis.

Longitudinal studies of mental health problems are needed to investigate relationships between mental health problems, such as conduct problems, anxiety, and sadness/depression (mental illness), and NDD, a mental disorder separate from mental illness.

#### 3.1.4. Mental Health Problems and Wellbeing (Mental Health) over the Life Course

Mental health problems vary over the life course with certain periods, such as adolescence, having both biological change and changes in life role expectations that increase the likelihood of mental health problems. Periods of more, or fewer, mental health problems exist in life for all people. We have used Halfon et al.'s [26] illustration, originally intended to describe changes in "health" over the life span, to visualize the life span trajectories of mental health problems and mental health (see Figure 2). Figure 2 illustrates that mental health problems can vary over time on a continuum from no problems to severe mental health problems. It is possible that neither complete wellbeing nor severe mental health problems/mental illness occur frequently. The same figure can be used to illustrate variations over the life span in mental health, defined as a state of wellbeing (see Table 1). When studying the trajectories of mental health problems in children with NDD, we are primarily interested in how mental health problems vary over time. Studying the occurrence of mental health problems may, however, not be enough—wellbeing is also important.

**Figure 2.** Wellbeing in a life span perspective, adapted from Halfon et al. (2014) [26].

#### *3.2. Problem 2: Distinguishing Mental Health Problems and Wellbeing as Separate Constructs in Children with NDD*

We propose that mental health should not be reduced to the absence of mental illness, but should encompass variations in mental health on a continuum from low to high levels of wellbeing.

Research in positive psychology and related fields have employed numerous conceptualizations of positive mental health and wellbeing [27,28]. Each understanding of the concept may present advantages and disadvantages, and arguments for each definition could be based on validity, pragmatic aspects, logic, and so forth. One characteristic of a construct that is sometimes overlooked is whether its definition is equally valid for the full width of human experience and functioning. If the overarching goal in wellbeing research is to describe universal as well as unique aspects of human functioning, then there is relatively little utility for concepts that are only valid for a subgroup of humanity, such as typically developed adults in western countries.

#### 3.2.1. Mental Health—A Multidimensional Wellbeing Concept

The WHO's definition of mental health [18] explicitly equates mental health with wellbeing. The WHO's definition can be used easily when working with adults without severe cognitive impairments. It is not as easy to apply to children and adolescents within the NDD spectrum, because their ability to meet aspects of the definition—"realizing abilities, cope with stress, work productively, and make a contribution to society"—may, by definition, preclude a determination of "wellbeing".

Usually, wellbeing is seen as comprising positive emotional states (feeling good) [29–32] and as having fewer/lower negative emotional states [31,33]. Some authors also describe good functioning as being a part of wellbeing [29,30], including having a command over resources or achieving a balance between resources and challenges [34]. In a study of student perspectives, wellbeing was found to be related to being (e.g., happy, satisfied), having (e.g., rights, relationships, resources, voice), and doing (e.g., looking after self and others, having goals, and making good decisions) [35]. The three dimensions of being, having, and doing can apply to all people, including children with NDD, and can be linked to two dominating, broad perspectives in wellbeing research: hedonia and eudaimonia [36]. Thus, wellbeing's positive emotional states include the two different ideas of happiness: hedonic (happiness or pleasure), that is living a pleasant life, or eudemonic (striving for, achieving something more—either personal growth or something outside the self), that

is, living a goal directed or meaningful life [19]. People experience both hedonic and eudemonic happiness but may seek or value one type of wellbeing more than the other. In children and adolescents with significant NDD, the "doing" and edudaimonic elements of wellbeing may have a restricted range of expressions or require substantial support from others; however, they are not by definition excluded from the experience.

It has been suggested that wellbeing may be best understood as a multidimensional phenomenon incorporating both ideas of wellbeing [36]. One attempt at combining hedonic and eudaimonic influences is seen in Keyes et al. [20] work. Keyes argues that mental health consists of three partly overlapping dimensions of wellbeing: emotional wellbeing (entailing positive affect, absence of negative affect, and perceived satisfaction with life), psychological wellbeing (consisting of self-acceptance, positive relations with others, personal growth, purpose in life, environmental mastery, and autonomy), and social wellbeing (social acceptance, social actualization, social contribution, social coherence, and social integration). When testing this suggestion, Keyes et al. [20] found support for a two-factor wellbeing model, corresponding to the two traditions: eudaimonia, comprising psychological and social wellbeing indicators, and hedoninia, comprising subjective (emotional) wellbeing.

#### 3.2.2. A Dual Model of Mental Health and Mental Health Problems

Because the WHO has provided both a definition of mental disorders and a definition of mental health in which mental health is explicitly named as wellbeing, the relationship between wellbeing and mental health problems needs clarifying. Do wellbeing and mental health problems exist on the same continuum? Literature describing wellbeing as the presence of positive feelings towards your own life tends to see wellbeing as a continuum of its own. Keyes [37] considers levels of wellbeing on a scale anchored by languishing (unhappiness and experiencing difficulties) at one end and flourishing (happy and thriving—the most positive state) at the other [37,38]. Several studies provide evidence that mental health problems and wellbeing are two separate but correlated constructs, rather than one (MacArthur Foundation's Midlife in the United States survey) [39]. Studies including children with NDD lend further support to this dual-factor model of mental health [40,41]. The term flourishing is suggested as useful to characterize people with high scores in emotional, psychological, and social wellbeing, whereas languishing can be used to categorize people with low scores on wellbeing. Thus, languishing is seen as indicating a low level of wellbeing that might, or might not, occur in conjunction with mental health problems or illness.

In conclusion, the support for the dual continua model means that we can add another layer to Figure 1. It is theoretically possible for someone to fulfill the criteria for a mental disorder (e.g., autism) and to also experience any level of mental health problems and wellbeing (circles partly overlapping). Mental illness most likely influences a person's wellbeing, but in theory, it is possible to experience aspects of positive mental health such as wellbeing when suffering from a mental health problem. The relationship between mental health problems and the dual continua model is illustrated in Figure 3.

**Figure 3.** The relation of mental disorders and mental illness to the dual-continua model of wellbeing and mental health problems.

#### *3.3. Problem 3: The Relationship between Participation and Wellbeing in Children with NDD*

We propose that participation is a key concept to relate to wellbeing because of its focus on functioning in the context of everyday activities. Participation can be an antecedent of wellbeing as well as a consequence of wellbeing.

In discussing the relationship between wellbeing and participation, we consider the antecedents and consequences of wellbeing and participation as described across diverse literature bases such as children with disabilities, aging populations, and the business literature. Antecedents and consequences provide information about possible causal links between the constructs of wellbeing and participation, although both wellbeing and participation are complexly determined and may have a cascade of effects. First, we provide descriptions of the construct of participation.

#### 3.3.1. How Participation Is Conceptualized in Various Bodies of Literature

Recent research in the child-onset disability field identifies participation as being involvement in a life situation with two dimensions. The first dimension attendance relates to the life situations and the second dimension to the involvement or engagement while being there. The dimensions are situated within the fPRC, which is neutral about the activity or life situation in which participation occurs, that is, participation can be considered in relation to any activity [5] and is pertinent for all people.

Although participation can occur in any life situation, the need to identify the situation in which participation is being studied implies that participation is a contextually based construct. Research about participation can be found in diverse literature, for example, business literature that focuses on participation in work (e.g., [42,43], or youth delinquency research that focuses on participation in crime or the legal system [44]. Some participation research implies that participation requires others to be present—thus effectively limiting the types of life situations in which participation can be said to have occurred. For example, in the aged care/adult disability literature, there has been a focus on participation being relevant in "social", "community", or "complex" activities [45,46]. The fPRC describes participation as being relevant to any life situation, including activities done individually, thus providing important conceptual clarity and applicability to all people.

Across various fields of literature are examples of studies in which the term participation is not defined explicitly: presumably based on the assumption that we all know and agree about what it is. When participation is not defined, what is measured is commonly the "attendance" dimension: that is, how often people attend particular activities, or what proportion of people attend particular situations. The notion of involvement is further explored here, because there is greater variation across literature on how involvement is operationalized compared to attendance.

Research about participation in decision making provides one mechanism for exploring involvement. Decision making is a process—whether done collaboratively or independently—and can be relevant to any life situation. Concerning participation in community development, the implied definition of participation is both attendance (in the decision-making activity) and involvement in dialogue [47]. This perspective is consistent with youth delinquency research studies in which participation in the legal proceedings has been considered in relation to involvement in decision making and problem solving around issues directly affecting the individual [48]. A focus on collaboration in decision making is also apparent in some education literature that describes participation as children being listened to by adults and having their views considered in decision making [49]. Puritz and Majd [50] describe involvement as having a meaningful opportunity to be heard.

Participation defined as "taking part", which might include interacting, doing, helping, or contributing [45], or as engagement in (complex) activities [46], also provides ideas about involvement. Operationalizing these ideas, however, often results in "counting occasions of doing (something)" an idea that is closer to the notion of attending than the experience of involvement. Likewise, in the business and education literature, although the term engagement is used more commonly than participation, the focus is frequently related to "engaged time" [51]—once again a measure of attendance. In contrast, engagement defined by Russell et al. [52] as "energy in action, the connection between person and activity" (p. 1), conveys the essence of the experience of participation, and reinforces the need to consider participation in context.

Bringing the ideas of attendance and involvement together, Bergqvist et al. [53] reported that "when a person chooses to attend an activity, it is possible for the person to be involved and that might lead to participation" (p. 1). In this example, participation is seen as a potential outcome of doing something, which suggests that participation cannot be separated from either doing or belonging. This definition of participation is consistent with the fPRC from the perspective that attendance is seen as a necessary but not sufficient condition for involvement.

Hoogsteen and Woodgate's [54] conceptual analysis of participation through the lens of childhood disability resulted in a definition of participation with four elements: "(i) the child must take part in something or with someone; (ii) the child must feel included or have a sense of inclusion in what they are partaking in; (iii) the child must have a choice or control over what they are taking part in; (iv) the child must work towards obtaining a personal or socially-meaningful goal or enhancing the quality of life" (pp. 329–330). The first two elements are consistent with the ideas of attendance and involvement. The third element is problematic as children often participate in activities or situations that they do not choose or control; however, the problematic nature of this element relates specifically to the attendance aspect of participation, as providing children with choices within an activity setting can help them feel involved or engaged [55,56]. This reflects an empowerment approach to the design of participation opportunities. The fourth element proposed seems closer to definitions of wellbeing than participation, but might point to the notion of future participation being driven by past and current participation—i.e., participation as a means.

#### 3.3.2. Antecedents and Consequences to Wellbeing and Participation

The relationship between participation and wellbeing must be considered as a transactional process over time where participation at one point in time may affect wellbeing at a later point in time, and vice versa. To further consider this relationship within a process framework, two concepts that denote a causal order of events will be used: antecedents and consequences. Antecedents concern events that occur before a specified event and consequences concern events that occur after a specified event.

Antecedents to wellbeing have been described as relating to resources or contextual factors and to personal attributes. For example, having social capital and enough income [57] can support wellbeing. From a personal perspective, altruism or volunteering [32,58] and adapting to your own needs for wellbeing and your life circumstances [29]

have all been identified as antecedents to wellbeing. Antecedents to low levels of wellbeing (i.e., languishing) may also be resources and contextual factors—for example, family and work variables and life stressors [32,59], limited resources [32], and lack of social engagement [60,61]. Of the factors identified as antecedents to wellbeing, participation is rarely explicitly described, although can be inferred from the literature describing altruistic behaviors, adaptive behaviors, and social engagement. Powell et al. [35] is one exception: they clearly identified participation as an antecedent contributing to wellbeing.

Consequences of wellbeing include protection against mental health problems [62], future resilience and wellbeing [57,63], connectedness with peers [57], improved work/school productivity, engagement and achievement [59], and a sense of meaning in life [28]. Thus, one consequence of wellbeing appears to be participation; other consequences relate to personal attributes of resilience, coping, and future wellbeing.

Antecedents of participation from across the fields of literature can also be summarized as factors related to the person or the context. Person-related antecedents include interest or willingness to take part [54], and past satisfaction [64], as well as antecedents that prevent participation, such as pain [64], depression, mood disorder [32], fatigue, or physical limitations [65]. Age was proposed to shape participation in that it influences capacity for choice. Contextual antecedents of participation included initiatives that influence the physical, attitudinal, and relational environment [33,42,43]; information provided [66]; and peer modelling, family processes, socioeconomic factors, cultural practices, and governance structures [67,68]. These broad-ranging antecedents provide information about how contexts might be shaped or influenced to support participation.

Consequences of participation as attendance included gaining skills, academic or educational achievement, health, development of self-determination or self-efficacy, overall development, and wellbeing [51,69–72]. The consequences of participation as involvement if seen as collaboration in action and decision making included impacts at the level of both person and context. For example, having agency or power and being able to contribute to choices that impact the future are personal consequences; societal transformation to realize rights and more equitable distribution of resources and benefits are contextual consequences [47,71]. Examples of consequences of participation in harmful activities were reported to include poor mental health, substance abuse, cynicism, and societal disengagement and crime [67,68]—again involving both personal and contextual consequences, strongly supporting the reciprocal nature of participation in context.

Consequences of a lack of participation were reported to include deprivation, social injustice, limited wellbeing, lack of dignity, loss of rights [47,66], and a lack of involvement leading to lack of attendance at work or low productivity [33]. A lack of participation can lead to a lack of contribution to building social capital by particular groups in society. For example, if those with disability are not participating, their potential to shape culture, build tolerance to diversity, benefit from and contribute to common resources, and establish valued norms impacts the nature of community/society for all [73]. Additionally, the consequences of imbalanced participation, i.e., not being able to achieve balance in doing all the activities that "need" to be done and resting, included stress and mental fatigue [53].

#### 3.3.3. Relationships between Participation and Wellbeing

The descriptions of wellbeing are primarily focused on the person's summative perception of their feelings about their life in terms of emotions, psychological functioning, and/or social wellbeing or a specific domain of life (e.g., recreation, work), whether focused on pleasure or striving or a combination. In contrast, descriptions of participation focus on the person taking part in context. In relation to the fPRC, wellbeing might be most closely related to ideas of "sense-of-self", which is described as both antecedent and consequent to participation in the fPRC. The broader literature related to participation also clearly (and commonly) links wellbeing as both an antecedent (when poor [i.e., when people are languishing] it limits/reduces participation) and a consequence of participation. When participation is possible, balanced, and not in harmful activities, wellbeing (flourishing)

can be enhanced. If participation is not balanced or is predominantly in harmful/negative activities, wellbeing is seen to reduce. Thus, participation and wellbeing are bi-directional: participation can influence wellbeing, and (positive) wellbeing can increase the possibility of participation [59].

Van Campen and Ledema [74] investigated the relationship between participation and wellbeing specifically, providing evidence about the need to understand both dimensions of participation. They focused on the impact of objective participation (attendance) on subjective wellbeing. They hypothesized a linear relationship between duration of illness leading to severity of impairment leading to objective participation leading to subjective wellbeing. Objective participation was measured as the frequency of hours in paid work, frequency of social contacts, number of holidays, and number of museum visits (thus measures of attendance). Subjective wellbeing was measured as health-related quality of life, using scales capturing mental health problems, and a measure of happiness (wellbeing). They found no empirical support for a direct relationship between objective participation and mental health problems or wellbeing. When models were adjusted to include age and socio-economic factors, a better fit was seen. In the discussion, the authors identified the need to understand subjective participation to understand its impact on wellbeing. They cited Csikszentmihalyi's notion of flow and interpreted this finding as follows: "it is not the fact that someone participates but how they participate that determines subjective wellbeing" (p. 643).

#### **4. Implications for Measurement and Intervention with Children with NDD Following from the Three Propositions**

The three problems discussed have implications for how mental health problems, wellbeing, and participation are measured in studies focusing on children with NDD. There are also implications for interventions focusing on decreasing mental health problems or enhancing wellbeing. Measurement and intervention are important topics that require consideration beyond the scope of this paper. In this section, we briefly point to some areas that need further discussion and empirical investigation.

#### *4.1. Implications for Measurement: The Risk of Confusion between NDD-Core Symptoms, Mental Health Problems, and Wellbeing*

One essential aspect of any instrument aiming to measure mental health problems or screen for mental illness in children with NDD is that it should not tap into core problems associated with the NDD in question. Looking at two of the most widely used behavior problem screening questionnaires for children and adolescents, the Child Behavior Checklist (CBCL) [24] and the Strengths and Difficulties questionnaire (SDQ) [75], it is apparent that both contain several items that risk doing so (e.g., "avoids looking others in the eye" from the CBCL and "easily distracted, concentration wanders" from the SDQ). This suggests that the problem of confusing NDD symptoms and mental health problems may apply to a substantial proportion of the research on mental health problems undertaken with children with NDDs.

This issue is equally important when measuring wellbeing, since the presence of an NDD does not predispose individuals to either languishing or flourishing. This problem does not primarily lay within the rating scales themselves but in how data are treated. For example, concerning mental health problems, the SDQ [75] is commonly used to screen mental health problems in children with NDD, e.g., Bailey et al. [22]. In the SDQ, there are four "problem scales": (i) hyperactivity (covering problems with both hyperactivity and inattention—the basic symptom criteria for Attention Deficit Hyperactivity Disorder), (ii) conduct problems, (iii) emotional problems (sadness, depression), and (iv) peer problems (problems in relating to peers). The subscales hyperactivity and peer problems should not be defined as mental health problems of an individual. Hyperactivity can exist along with good everyday functioning as operationalized as participation in play activities in preschool [76]. Peer problems are related to how other people react to a child and the child's communication skills; thus, this scale is also a measure of communicative

and environmental problems. For this reason, we recommend caution when drawing conclusions based on indexes, such as the internalizing or externalizing indices of the SDQ and CBCL, about mental health problems in children and adolescents with NDD diagnoses.

#### *4.2. Implications for Measurement: The Issue of Inclusiveness*

A related and equally important aspect of measurement instruments is the matter of inclusiveness at the conceptual level, that is, items and scales should not preclude any level of wellbeing or mental health problems based on normative assumptions of human functioning (it should be a purely empirical question). For example, if working "productively and fruitfully" is considered by WHO [18] as a central part of wellbeing, then the individuals with the severest disabilities, for whom work in the traditional sense will never be an option, are predestined to lower levels of wellbeing. One way of reducing the risk of building conceptual barriers may be to let respondents assess wellbeing in general with a few items or using a single question. There are of course limitations to such approaches that reduce a complex phenomenon to a few items. We recommend that researchers and clinicians consider the inclusiveness of any scale chosen to measure wellbeing and mental health problems in children with NDD. Given our definition of wellbeing as subjective, we realize this recommendation is difficult to follow in the case of individuals with profound intellectual disabilities. This literature tends to use proxycompleted measures of quality of life (not wellbeing), such as the KidsLife Scale [77], which is based on a series of life domains including self-determination, social inclusion, and interpersonal relationships, in addition to material, physical, and emotional wellbeing.

Even after having considered the risk of confusing mental health problems with core symptoms of NDD and inclusiveness, the questions of inclusive measurement design and procedures remain. Many questionnaires have cognitive barriers that may make them inaccessible for children with NDDs. Instruments suited for assessing mental health problems and wellbeing in children with NDD need to be developed or adapted. In addition, manuals for how to set up structured interviews to support individuals in selfrating wellbeing and mental health problems need to be developed. One example of a questionnaire that tries to deal with these issues constructively is the Wellbeing in Special Education Questionnaire [40]. The instrument has been validated with children with mild to moderate intellectual disability and includes generic questions about wellbeing along with questions about mental health problems that could be argued to be relevant for children regardless of the level of functioning.

Conceptual inclusiveness is also pertinent to measures of participation. Following the publication of the International Classification of Functioning, Disability, and Health [6], in which the concept participation was introduced, multiple participation measures were developed. However, the lack of conceptual clarity within the ICF led to considerable variation in approaches to measurement development [78]. One of the key issues was the conflating of the ideas of independence in performing an activity or task, with involvement in life situations (the ICF definition of participation). The problem with this approach is that children with NDD were, by definition, assessed as having poor or restricted participation simply because they were not independent (e.g., they required supports to participate due to intellectual impairment) or had limitations in their activity skills (e.g., poor manual ability). In terms of measuring participation, the inclusion of an assessment of support or aids required to participate has been critiqued in the literature [79,80]. It is considered important to conceptualize participation intrinsically and separately from other factors or variables [79]. Children with NDD may experience participation restrictions, but this should not be determined based on their skills, or attributes associated with their condition [5]. Participation attendance (being there) and involvement (the experience of participation while attending) in life situations are pertinent to all people at all phases of the life course. Measures of participation should reflect one or both these constructs.

#### *4.3. Interventions Focused on Decreasing Mental Health Problems in Persons with NDD*

Interventions that address mental health problems with anxiety and sadness/ depression in persons with NDD are limited. Pharmacological interventions for severe mental health problems, such as antidepressant medication, may not be effective [81]. Nonpharmacological interventions have focused on talking therapies. Mindfulness (combining talking with meditation) has been shown to be effective for reducing anxiety in persons with autism [82] and a cognitive–behavioral therapy combination of on-line sessions and face to face meetings has been shown to reduce anxiety in adolescents with intellectual disability [83]. Evidence for the effect of psychotherapy is primarily limited to case-studies [84]. There is emerging evidence that talking therapies need to be modified for young people with NDD [85]. There is a dearth of evidence for talking therapies for persons with NDD who may experience more significant motor and communication difficulties. Few studies focusing on decreasing mental health problems measure wellbeing or participation of children and youth with NDD as secondary outcomes of treatment.

#### *4.4. Participation Interventions as a Means to Enhancing Wellbeing in Children and Adolescents with NDD*

The childhood disability literature is just beginning to explore the effects of participation interventions on wellbeing. Studies of various participation interventions, including arts-based, physical activity, life skills, coaching, and resilience-focused interventions, have provided preliminary evidence for effects on wellbeing (e.g., psychosocial well-being, self-determination, self-efficacy). For example, a scoping review of arts-based interventions for children with disabilities, which included performance (e.g., music, dance, theatre) and visual (e.g., drawing, painting, sculpting) arts-based programs, indicated that these interventions show potential to positively impact psychosocial wellbeing (i.e., emotional and social functioning), although further investigation is required with broader populations of children with physical and developmental disabilities [86]. Therapeutic horse riding, an example of a physical activity intervention, has been found to positively influence and expand the self-concepts of children with disabilities [87]. A review of the literature on therapeutic horseback riding indicates some evidence for statistically significant decreases in depression and distress, although this evidence is inconsistent and there are methodological problems in this body of research [88]. Youth with disabilities taking part in a transition-oriented life skills program have been found to have significant pre-post changes in their autonomy (as an aspect of self-determination) and self-efficacy [56]. The growing literature on coaching interventions for children and youth with disabilities focuses on engagement and goal attainment [89,90] and has yet to consider longer-term effects on wellbeing. However, the broader coaching literature indicates that participation in a cognitive–behavioral life coaching program is associated with enhanced wellbeing and quality of life [91]. Resilience-focused interventions are another promising area of intervention. A systematic review of universal resilience-focused interventions targeting child and youth wellbeing in the school setting [92] has indicated effects concerning the reduction of mental health problems.

#### **5. Conclusions**

This position paper suggests future directions in the scientific study of wellbeing and mental health problems in children with NDD and describes the implications for participation interventions aimed at sustaining wellbeing in children with disabilities following from the propositions:

(1) Mental disorders include both diagnoses related to impairments in the developmental period, i.e., NDD and diagnoses related to mental illness. These two types of mental disorders must be separated when measuring mental health in children with disabilities. Thus, summary indexes such as externalizing and internalizing problems should be avoided, since more stable characteristics related to impairment are conflated with mental health problem indicators. Measures of mental health problems involving

only mental illness indicators and not NDD impairment-related symptoms need to be developed for children diagnosed within the NDD spectrum.


**Author Contributions:** M.G., C.I., G.K., A.K.A., L.A. (Lilly Augustine), R.B., H.D., J.G., M.I., L.-O.L., F.L. and L.A. (Lena Almqvist) contributed to the conceptualization, writing the original draft, and the review and editing of the manuscript. M.G. had a leading role in the conceptualization, funding acquisition, writing the original draft, and the review and editing of the manuscript. C.I., G.K., and L.A. (Lena Almqvist) have also had leading roles in the review and editing of the manuscript. The visualizations were adapted (Figures 1 and 2) and created (Figure 3) by M.I. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research took place within the research program CHILD-PMH (Child Participation and Mental Health), which was funded by the Swedish Research Council (VR) grant number 2018-05824.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** This conceptual paper was facilitated by the administrative support of Helena Engkvist and by creative discussions with the rest of the researchers involved in CHILD-PMH.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


International Journal of *Environmental Research and Public Health*
